diff --git a/.claude/INVARIANTS.md b/.claude/INVARIANTS.md
new file mode 100644
index 000000000..698282a0c
--- /dev/null
+++ b/.claude/INVARIANTS.md
@@ -0,0 +1,233 @@
+---
+title: Invariants Index
+purpose: constrain
+tier: stable
+audience: agent
+status: active
+topic: [ax, invariants]
+last-verified: 2026-05-16
+---
+
+# VariScout Invariants
+
+Cross-cutting non-negotiables. Each invariant has ONE canonical home (linked) + ONE enforcement mechanism (lint rule, structural-absence test, denylist guard, or convention + code review). Subagents loading the `agent-context-quickstart` skill see this doc on session start.
+
+Enforcement tiers:
+- **ESLint** — blocks at lint-time; CI fails if violated.
+- **Architecture test** — structural-absence guard in `packages/*/src/__tests__/`; CI fails if violated.
+- **Denylist script** — bash script run in pre-commit / `bash scripts/pr-ready-check.sh`; blocks push.
+- **Convention** — no mechanical gate; reviewer-enforced + code review protocol.
+
+---
+
+## Hard Invariants (lint- or test-enforced)
+
+### Language
+
+- **Never "root cause"** — say "contribution" / "suspected cause" / "mechanism" (P5 amended).
+  - Canonical: [ADR-073](../docs/07-decisions/adr-073-no-statistical-rollup-across-heterogeneous-units.md) §Rationale + [`.claude/invariants/coscout-prompts.md`](invariants/coscout-prompts.md).
+  - Enforced: ESLint `variscout/no-root-cause-language` (scoped to AI prompts + CoScout code).
+
+- **Interactions are `'ordinal'` / `'disordinal'` — never "moderator" or "primary"**.
+  - Canonical: [`.claude/invariants/stats.md`](invariants/stats.md).
+  - Enforced: ESLint `variscout/no-interaction-moderator` (scoped to stats + regression code).
+
+### Stats engine
+
+- **Never return `NaN` or `Infinity`** — return `number | undefined`; use `safeMath.ts` (`finiteOrUndefined`, `safeDivide`, `computeOptimum`).
+  - Canonical: [`.claude/invariants/stats.md`](invariants/stats.md).
+  - Enforced: convention + three-boundary numeric safety (ADR-069).
+
+- **Never `.toFixed()` on stat outputs** — display via `formatStatistic()` from `@variscout/core/i18n`.
+  - Canonical: [`.claude/invariants/stats.md`](invariants/stats.md) + [`.claude/invariants/i18n.md`](invariants/i18n.md).
+  - Enforced: ESLint `variscout/no-tofixed-on-stats`.
+
+- **Never `Math.random()` in any code or test** — tests use seeded PRNG helpers; production code avoids randomness.
+  - Canonical: [`.claude/invariants/stats.md`](invariants/stats.md) + [`.claude/invariants/testing.md`](invariants/testing.md).
+  - Enforced: convention + code review (no ESLint rule currently — see investigations.md for open gap).
+
+- **NIST Longley fixture must stay green to 9 significant digits** — never weaken the threshold.
+  - Canonical: [`.claude/invariants/stats.md`](invariants/stats.md).
+  - Enforced: Vitest test fixture (failing it fails CI).
+
+- **No cross-investigation statistical aggregation** — no `meanCpk`/`sumCpk`/`portfolioCpk` across heterogeneous units. Distributions, not aggregates. Side-by-side comparison (facets) is always safe; arithmetic across heterogeneous physics is always wrong.
+  - Canonical: [ADR-073](../docs/07-decisions/adr-073-no-statistical-rollup-across-heterogeneous-units.md).
+  - Enforced: architecture test `packages/core/src/__tests__/architecture.noCrossInvestigationAggregation.test.ts`.
+
+### Chart components
+
+- **Colors only via `chartColors` / `chromeColors`** from `@variscout/charts/colors` — no hex literals.
+  - Canonical: [`.claude/invariants/charts.md`](invariants/charts.md).
+  - Enforced: ESLint `variscout/no-hardcoded-chart-colors`.
+
+- **LTTB must force-include UCL/LCL violations** — silent dropping is a correctness bug.
+  - Canonical: [`.claude/invariants/charts.md`](invariants/charts.md).
+  - Enforced: convention + code review.
+
+### ADR-074 level-spanning surface boundary
+
+- **Owner surfaces own their primitives** — Investigation Wall does not reimplement L1 chart rendering; SCOUT does not reimplement Evidence Map factor-network; FRAME does not embed hypothesis canvas surfaces; Evidence Map does not reimplement L2 flow rendering.
+  - Canonical: [ADR-074](../docs/07-decisions/adr-074-scout-level-spanning-surface-boundary-policy.md).
+  - Enforced: `scripts/check-level-boundaries.sh` (runs in `bash scripts/pr-ready-check.sh`).
+
+### Persistence boundary (ADR-059 / ADR-079)
+
+- **No direct Dexie import outside persistence layers** — only `apps/*/src/persistence/` and `apps/*/src/db/` may import Dexie directly. Exception: R12 (`packages/stores/src/canvasViewportStore.ts`), R13 sustainment direct writes with ESLint allow-list.
+  - Canonical: [ADR-059](../docs/07-decisions/adr-059-web-first-deployment-architecture.md) + F1+F2 P7.2 enforcement.
+  - Enforced: ESLint `no-restricted-imports` (P7.2 rule in `eslint.config.js`, R12+R13 allow-list).
+
+### CoScout prompts
+
+- **Entry point is `assembleCoScoutPrompt()`** — `buildCoScoutSystemPrompt()` in `legacy.ts` is deprecated (test backward-compat only).
+  - Canonical: [`.claude/invariants/coscout-prompts.md`](invariants/coscout-prompts.md).
+  - Enforced: convention + code review.
+
+- **Every tool in `tools/registry.ts` declares `phases` + optional `tier: 'team'`** — ungated tools leak across phases/tiers.
+  - Canonical: [`.claude/invariants/coscout-prompts.md`](invariants/coscout-prompts.md).
+  - Enforced: convention + code review.
+
+- **Tier 1 is session-invariant** — moving content tier1 ↔ tier3 breaks prompt-cache hit rate.
+  - Canonical: [`.claude/invariants/coscout-prompts.md`](invariants/coscout-prompts.md).
+  - Enforced: convention + code review.
+
+- **REF markers reference chart elements, not raw data values** (ADR-057, customer-owned data).
+  - Canonical: [`.claude/invariants/coscout-prompts.md`](invariants/coscout-prompts.md).
+  - Enforced: convention + code review.
+
+### i18n
+
+- **No string concatenation for translated content** — use `formatMessage()` with parameters; Intl APIs for plurals/numbers/dates.
+  - Canonical: [`.claude/invariants/i18n.md`](invariants/i18n.md).
+  - Enforced: convention + code review.
+
+- **`registerLocaleLoaders()` before any `preloadLocale()`** — apps call at startup; tests register their own loader before `beforeAll`.
+  - Canonical: [`.claude/invariants/i18n.md`](invariants/i18n.md).
+  - Enforced: convention + test failures (locales silently fall back to English without registration).
+
+### Testing
+
+- **`vi.mock()` BEFORE component imports** — mocks after imports cause infinite loops.
+  - Canonical: [`.claude/invariants/testing.md`](invariants/testing.md).
+  - Enforced: convention (test hangs surface the violation).
+
+- **`vi.mock()` factories referencing `@variscout/core` exports use `importOriginal` partial-pattern** — flat factories crash on transitive `DEFAULT_TIME_LENS` reads.
+  - Canonical: [`.claude/invariants/testing.md`](invariants/testing.md).
+  - Enforced: convention + test failures.
+
+- **Floats: `toBeCloseTo(value, decimals)`** — never `toBe()` for float comparisons.
+  - Canonical: [`.claude/invariants/testing.md`](invariants/testing.md).
+  - Enforced: convention.
+
+- **E2E selectors: `data-testid` only** — text/role/class selectors break with i18n + theme changes.
+  - Canonical: [`.claude/invariants/testing.md`](invariants/testing.md).
+  - Enforced: convention + code review.
+
+- **Zustand stores: reset state in `beforeEach`** via `useStore.setState(useStore.getInitialState())`.
+  - Canonical: [`.claude/invariants/testing.md`](invariants/testing.md).
+  - Enforced: convention.
+
+---
+
+## Soft Invariants (convention + reviewer-enforced)
+
+### Product data
+
+- **Browser-only processing** — data stays in customer's Azure tenant. No server-side aggregation; no external API calls with row data.
+  - Canonical: [ADR-059](../docs/07-decisions/adr-059-web-first-deployment-architecture.md).
+  - Enforced: code review (no server-processing entry points in product code).
+
+### Azure auth + storage
+
+- **EasyAuth only** — no MSAL in client code; `/.auth/me` returns identity.
+  - Canonical: [`.claude/invariants/azure-storage.md`](invariants/azure-storage.md).
+  - Enforced: convention + code review.
+
+- **No PII in App Insights** — log structural events only (counts, types, durations); send error *type*, not message text.
+  - Canonical: [`.claude/invariants/azure-storage.md`](invariants/azure-storage.md).
+  - Enforced: convention + code review.
+
+- **SAS tokens minted server-side** via `/api/storage-token` — client never sees storage keys; container-scoped, 1h expiry.
+  - Canonical: [`.claude/invariants/azure-storage.md`](invariants/azure-storage.md).
+  - Enforced: convention + code review.
+
+- **ETag concurrency on hub-blob writes (ADR-079)** — callers must handle the `{ ok: false; reason }` branch.
+  - Canonical: [`.claude/invariants/azure-storage.md`](invariants/azure-storage.md) + [ADR-079](../docs/07-decisions/adr-079-hub-blob-storage-etag-concurrency.md).
+  - Enforced: TypeScript typed result (compile-time guard); convention for branch handling.
+
+### Architecture shape
+
+- **6 Zustand stores across 3 layers** — Document (×3): `useProjectStore`, `useInvestigationStore`, `useCanvasStore`; Annotation (×2): `useCanvasViewportStore`, `usePreferencesStore`, `useActiveIPStore`; View (×1): `useViewStore`. No DataContext.
+  - Canonical: [ADR-078](../docs/07-decisions/adr-078-pwa-azure-architecture-alignment.md) + `packages/stores/CLAUDE.md`.
+  - Enforced: `packages/stores/src/__tests__/layerBoundary.test.ts` (structural-absence guard on layer names).
+
+- **Package dependencies flow downward** — `core → hooks → ui → apps`. Never import upward.
+  - Canonical: root `CLAUDE.md` §Invariants + [ADR-078](../docs/07-decisions/adr-078-pwa-azure-architecture-alignment.md).
+  - Enforced: convention + ESLint (some packages); TypeScript compiler catches circular deps.
+
+- **Sub-path exports need BOTH `package.json#exports` AND `tsconfig.json#paths` updated together** — updating one but not the other silently breaks imports.
+  - Canonical: root `CLAUDE.md` §Invariants.
+  - Enforced: convention + build failures (`pnpm build` catches type gaps).
+
+- **`@source` directive in every `apps/*/src/index.css` for shared packages** — missing it silently breaks Tailwind v4 responsive utilities.
+  - Canonical: root `CLAUDE.md` §Invariants.
+  - Enforced: convention (visual regression surfaces it; see `feedback_verify_before_push`).
+
+- **No bare `useStore()` calls** — always use selectors: `useProjectStore(s => s.field)`.
+  - Canonical: `packages/stores/CLAUDE.md`.
+  - Enforced: convention + code review.
+
+- **Strategy pattern for modes** — `resolveMode()` + `getStrategy()` is the sole source of truth for mode-specific behavior. No cascading mode ternaries.
+  - Canonical: [Constitution §10](../docs/01-vision/constitution.md) + `packages/core/CLAUDE.md`.
+  - Enforced: convention + code review.
+
+### Chart components (soft)
+
+- **Theme via `useChartTheme()`** — never read `data-theme` directly.
+  - Canonical: [`.claude/invariants/charts.md`](invariants/charts.md).
+  - Enforced: convention + code review.
+
+- **Pair `text-{color}-400` with `text-{color}-700`** for label contrast — `red/amber/green-400` fail light-mode contrast alone.
+  - Canonical: [`.claude/invariants/charts.md`](invariants/charts.md).
+  - Enforced: convention + visual regression (see `feedback_green_400_light_contrast`).
+
+- **No manual `React.memo()` on new chart components** — React Compiler handles memoization.
+  - Canonical: [`.claude/invariants/charts.md`](invariants/charts.md).
+  - Enforced: convention + code review.
+
+### Git / workflow
+
+- **Never `--no-verify` on commits** — hooks enforce invariants; bypassing hides real violations.
+  - Canonical: root `CLAUDE.md` §Workflow + `feedback_subagent_no_verify`.
+  - Enforced: convention; reviewers reject `--no-verify` evidence in PRs.
+
+---
+
+## Topic-Scoped Invariants (load on demand)
+
+Load these when working in the relevant area — they provide more detail than the summaries above.
+
+| Topic                          | Invariant file                                                   | When to load                                        |
+| ------------------------------ | ---------------------------------------------------------------- | --------------------------------------------------- |
+| Stats engine + math            | [invariants/stats.md](invariants/stats.md)                       | Writing or modifying stats/findings code            |
+| Test code (Vitest + Playwright)| [invariants/testing.md](invariants/testing.md)                   | Writing or modifying any test file                  |
+| Chart components               | [invariants/charts.md](invariants/charts.md)                     | Chart component work                                |
+| CoScout AI prompts             | [invariants/coscout-prompts.md](invariants/coscout-prompts.md)   | Editing CoScout prompts / AI tools / prompt registry. For CoScout work: load [coscout-ax-design](../docs/01-vision/coscout-ax-design.md) first (canonical AX design: persona, tier-gating, knowledge architecture, safety, eval), then this invariant file for prompt engineering details. |
+| i18n strings + locale          | [invariants/i18n.md](invariants/i18n.md)                         | Adding or modifying localized strings               |
+| Azure storage + auth           | [invariants/azure-storage.md](invariants/azure-storage.md)       | Azure Blob/Dexie sync, auth, SAS token work         |
+
+---
+
+## Enforcement Gaps (open)
+
+The following soft invariants lack a mechanical enforcement mechanism. Gaps are logged in `docs/investigations.md`.
+
+- `Math.random()` ban: enforced by convention only; no ESLint rule. A `no-math-random` rule scoped to `packages/core/src/**` + `**/__tests__/**` would close this.
+- `@source` directives: no build-time check; only visual regression surfaces it. A `pnpm docs:gen-arch`-style check for Tailwind source maps could close this (Play 7 candidate).
+
+---
+
+## Related
+
+- [docs/agent-context/onboarding-quick-start.md](../docs/agent-context/onboarding-quick-start.md) — 5-min agent orientation that summarizes the 4 hard invariants
+- [docs/agent-context/package-router.md](../docs/agent-context/package-router.md) — routing table pointing to relevant invariant files per work area
+- [docs/01-vision/constitution.md](../docs/01-vision/constitution.md) — 10 product principles (overlaps with soft invariants above; constitution is canonical for P1–P10)
diff --git a/.claude/rules/azure-storage.md b/.claude/invariants/azure-storage.md
similarity index 100%
rename from .claude/rules/azure-storage.md
rename to .claude/invariants/azure-storage.md
diff --git a/.claude/rules/charts.md b/.claude/invariants/charts.md
similarity index 100%
rename from .claude/rules/charts.md
rename to .claude/invariants/charts.md
diff --git a/.claude/rules/coscout-prompts.md b/.claude/invariants/coscout-prompts.md
similarity index 80%
rename from .claude/rules/coscout-prompts.md
rename to .claude/invariants/coscout-prompts.md
index 62cb98795..ecbc5ba90 100644
--- a/.claude/rules/coscout-prompts.md
+++ b/.claude/invariants/coscout-prompts.md
@@ -3,6 +3,7 @@ paths:
   - "packages/core/src/ai/prompts/**"
   - "packages/core/src/ai/actionTools.ts"
   - "packages/core/src/ai/**"
+related: [coscout-ax-design]
 ---
 
 # CoScout prompts — non-negotiables
@@ -14,3 +15,5 @@ paths:
 - **Tier 1 stays session-invariant** — moving content tier1 ↔ tier3 breaks prompt-cache hit rate.
 
 Detailed architecture + tier model + mode coaching: `packages/core/CLAUDE.md`.
+
+See also: [coscout-ax-design](../../docs/01-vision/coscout-ax-design.md) — canonical AX-design surface (persona, tier-gating, knowledge architecture, safety, eval discipline).
diff --git a/.claude/rules/i18n.md b/.claude/invariants/i18n.md
similarity index 100%
rename from .claude/rules/i18n.md
rename to .claude/invariants/i18n.md
diff --git a/.claude/rules/stats.md b/.claude/invariants/stats.md
similarity index 100%
rename from .claude/rules/stats.md
rename to .claude/invariants/stats.md
diff --git a/.claude/rules/testing.md b/.claude/invariants/testing.md
similarity index 100%
rename from .claude/rules/testing.md
rename to .claude/invariants/testing.md
diff --git a/.claude/skills/agent-context-quickstart/SKILL.md b/.claude/skills/agent-context-quickstart/SKILL.md
new file mode 100644
index 000000000..276fa9253
--- /dev/null
+++ b/.claude/skills/agent-context-quickstart/SKILL.md
@@ -0,0 +1,59 @@
+---
+name: "Agent Context Quickstart"
+description: "Use at the start of any subagent dispatch to load a 5-minute VariScout orientation: current strategic direction (wedge V1), hard invariants, where to look, and common pitfalls. Use when onboarding a fresh subagent, starting a new session, or any time a subagent needs baseline project context before touching code or docs."
+---
+
+# Agent Context Quickstart
+
+## When to Use This Skill
+
+Use at the start of any subagent dispatch — especially:
+- First action of a fresh session or a freshly dispatched subagent
+- Before any code or doc edit in an unfamiliar area
+- When a task touches multiple packages or layers and orientation matters
+
+Do NOT skip this for "quick" tasks. The most common subagent errors (wrong store layer, root-cause language, hex colors, bare useStore()) stem from missing this context.
+
+## What This Skill Does
+
+Loads a 5-minute orientation document covering:
+1. What VariScout is and its current strategic direction (wedge pivot 2026-05-16)
+2. Hard invariants that must never be violated
+3. Where to look for specs, decisions, and package context
+4. Common pitfalls that cause ESLint failures and test breaks
+
+## How to Use
+
+1. Read the onboarding doc at `docs/agent-context/onboarding-quick-start.md` (in the repo root).
+2. Note the 6-tab nav (`Home · Projects · Process · Analyze · Investigation · Report`) and that "Improve" is a stage inside Projects, not a tab. (Verify against any recent amendments in `docs/decision-log.md` — the wedge area is actively evolving.)
+3. Read `.claude/INVARIANTS.md` for the full invariant index — canonical homes + enforcement mechanisms for every hard and soft rule. The onboarding doc §Hard Invariants gives the quick summary; INVARIANTS.md gives the authoritative detail.
+4. **If your task involves editing ANY canonical doc** (design spec, ADR, decision-log, etc.): read `docs/agent-context/doc-discipline.md`. It defines the SSoT-by-doc-type rules — design specs edit-in-place, ADRs use amendment-blocks-at-bottom, decision-log appends. Anti-pattern: creating `*-amendment-*.md` side files (HARD-FAILed by validator when Play 2b ships).
+5. Scan the Common Pitfalls section in the onboarding doc — check each against your planned edits before proceeding.
+6. If your task involves a specific package, invoke the `package-router` skill next.
+7. If your task reads or writes store state, invoke the `store-state-glossary` skill next.
+
+## Key Facts (memorize these)
+
+**Current direction**: Wedge V1 — single-product, single SKU (€99/mo), single persona (improvement specialist). Canonical: ADR-082 + wedge spec (`docs/superpowers/specs/2026-05-16-wedge-architecture-design.md`).
+
+**6 tabs**: `Home · Projects · Process · Analyze · Investigation · Report`. Improve = stage inside Projects. Many older docs predate this; treat as historical.
+
+**Package dependency direction**: `core → hooks → ui → apps`. Never import upward.
+
+**Hard stops** (all ESLint-enforced):
+- Never `Math.random()` in any code or test
+- Never hardcoded hex colors in charts
+- Never `--no-verify` on commits
+- Never "root cause" language (use "contribution" / "suspected cause" / "mechanism")
+- Never `'ordinal' | 'disordinal'` confusion — never call interactions "moderator" or "primary"
+
+## Supporting Documents
+
+- Full 5-minute onboarding: `docs/agent-context/onboarding-quick-start.md`
+- Invariants index: `.claude/INVARIANTS.md`
+- **Doc-update discipline (when editing any canonical doc): `docs/agent-context/doc-discipline.md`**
+
+Related skills:
+- `package-router` — which package CLAUDE.md to load for your work area
+- `store-state-glossary` — what state lives in which Zustand store
+- `writing-tests` — Vitest + RTL + Playwright test patterns
diff --git a/.claude/skills/package-router/SKILL.md b/.claude/skills/package-router/SKILL.md
new file mode 100644
index 000000000..6c0d3a676
--- /dev/null
+++ b/.claude/skills/package-router/SKILL.md
@@ -0,0 +1,57 @@
+---
+name: "Package Router"
+description: "Use when you need to figure out which package's CLAUDE.md, ADRs, rules, and skills to load for a specific work area. Covers: canvas, stats engine, chart components, stores, i18n, Azure-specific (storage/auth/sync), PWA-specific (persistence/Dexie), UI primitives, hooks, CoScout prompts, response paths, capability mode, performance mode, yamazumi mode, defect mode, process-flow mode, projects tab, Investigation Wall, Evidence Map. Use before starting any implementation in an unfamiliar area."
+---
+
+# Package Router
+
+## When to Use This Skill
+
+Use when you know *what* you need to do but not *where* to look. Given a work area (e.g., "I need to add a chart", "I need to touch the stores", "I need to fix a canvas bug"), this skill routes you to the right CLAUDE.md, ADRs, and rules before you write a line of code.
+
+## How to Use
+
+1. Find your work area in the routing table at `docs/agent-context/package-router.md`.
+2. Read the **Primary CLAUDE.md** in full first.
+3. Skim **Related CLAUDE.mds** for boundary rules relevant to your change.
+4. Look up the cited **ADRs** in `docs/07-decisions/` when you need the rationale.
+5. Check `.claude/rules/` files listed in the Rules column for machine-enforced constraints.
+6. Invoke any suggested skill.
+7. Check `docs/decision-log.md` before reopening any decided question.
+
+## Quick Reference (Top Work Areas)
+
+| Work Area | Primary CLAUDE.md | Key Constraint |
+|---|---|---|
+| Canvas / FRAME / Process Hub | `packages/ui/CLAUDE.md` | `Canvas` is canonical; `LayeredProcessViewWithCapability` deprecated |
+| Stats engine | `packages/core/CLAUDE.md` | `safeMath.ts`; return `undefined` not NaN; never `Math.random()` |
+| Chart components | `packages/charts/CLAUDE.md` | `chartColors` only; `useChartTheme()`; no hex literals |
+| Zustand stores | `packages/stores/CLAUDE.md` | Selector required; 3-layer boundary enforced |
+| i18n | `packages/core/CLAUDE.md` | `registerLocaleLoaders()` before `preloadLocale()` |
+| Azure (storage/auth/sync) | `apps/azure/CLAUDE.md` | EasyAuth only; no PII in telemetry; ETag concurrency |
+| PWA (persistence/Dexie) | `apps/pwa/CLAUDE.md` | Session-only by default; no AI in free tier |
+| UI primitives | `packages/ui/CLAUDE.md` | `*Base` = primitive; `*WrapperBase` = composition |
+| React hooks | `packages/hooks/CLAUDE.md` | Consume stores; never define stores |
+| CoScout prompts | `packages/core/CLAUDE.md` | `assembleCoScoutPrompt()`; tier 1 is cache-invariant |
+| Projects tab | wedge spec + `apps/azure/CLAUDE.md` | 4 stages: Charter→Approach→Improve→Sustainment |
+| Investigation Wall | `packages/core/CLAUDE.md` + `packages/stores/CLAUDE.md` | `FindingSource` discriminated union; no cross-hub aggregation |
+
+## Full Routing Table
+
+The complete table (20 rows with ADRs, rules, and suggested skills per work area) is at:
+
+`docs/agent-context/package-router.md`
+
+Read that doc for the authoritative routing matrix. The quick reference above covers the 80% case.
+
+## Package Dependency Invariant
+
+`core → hooks → ui → apps` — import direction is one-way downward. If your change requires importing upward, you need a different design.
+
+## Sub-path Export Rule
+
+Updating a sub-path in any package requires BOTH `package.json#exports` AND `tsconfig.json#paths` updated together. Missing either causes silent runtime failures.
+
+## Tailwind v4 Rule
+
+Every `apps/*/src/index.css` needs `@source` for each shared package that provides UI (`@source "../../../packages/ui/src/**/*.tsx"`, etc.). Missing `@source` silently breaks responsive utilities.
diff --git a/.claude/skills/store-state-glossary/SKILL.md b/.claude/skills/store-state-glossary/SKILL.md
new file mode 100644
index 000000000..e63aacef5
--- /dev/null
+++ b/.claude/skills/store-state-glossary/SKILL.md
@@ -0,0 +1,97 @@
+---
+name: "Store State Glossary"
+description: "Use when you need to know what state lives in which Zustand store before reading or mutating store state. Covers the 7 stores across 3 layers (Document ×3, Annotation ×3, View ×1) per ADR-078 + F4. Use before any store read/write, before dispatching actions, or when you're unsure which store owns a piece of state."
+---
+
+# Store State Glossary
+
+## When to Use This Skill
+
+Use before:
+- Reading or writing any Zustand store field
+- Dispatching a store action
+- Deciding where new state should live (which layer + which store)
+- Writing tests that reset store state in `beforeEach`
+- Debugging state that isn't persisting / is persisting unexpectedly
+
+## The 3-Layer Model
+
+| Layer | Stores | Persistence | Test reset |
+|---|---|---|---|
+| **Document** | `useProjectStore`, `useInvestigationStore`, `useCanvasStore` | Hub repository → `.vrs` / Dexie (Azure) | Load a fresh `SerializedProject` |
+| **Annotation** | `useCanvasViewportStore`, `usePreferencesStore`, `useActiveIPStore` | Dexie / idb-keyval / localStorage | Clear individually per store key |
+| **View** | `useViewStore` | NONE — transient | `useViewStore.setState(useViewStore.getInitialState())` |
+
+**Portability test**: Would another analyst importing this hub need this state? Yes → Document. Survives reload but not portable → Annotation. Doesn't survive reload → View.
+
+## Quick Field Reference
+
+### Document — `useProjectStore` (dataset + config)
+- `rawData` — all parsed rows
+- `outcome` — selected Y column
+- `factors` — selected X columns
+- `specs` / `measureSpecs` — spec limits (per-outcome and per-column)
+- `analysisMode` — `'standard' | 'performance' | 'yamazumi' | 'defect'` (never `'capability'` — that's a `ResolvedMode`)
+- `filters` — active row-level filters
+- `subgroupConfig` — n for subgrouping (default n=5+ for capability)
+- `dataQualityReport` — parse-time report
+
+### Document — `useInvestigationStore` (findings + hypotheses)
+- `findings` — pinned observations (`FindingSource` discriminated union; always narrow before accessing)
+- `questions` — investigation question tree (max depth 3, max 8 children per parent)
+- `hypotheses` — `SuspectedCause` entities; `selectedForImprovement: true` triggers HMW brainstorm
+- `causalLinks` — directed causal graph edges (cycle-guarded via `wouldCreateCycle()`)
+- `problemContributionTree` — AND-gate tree for the Wall (highlights in `useViewStore`)
+
+### Document — `useCanvasStore` (FRAME / Process Hub canvas)
+- `canonicalMap` — the process map (nodes = steps, edges = connections)
+- `outcomes` / `primaryScopeDimensions` — hub outcome specs
+- `undoStack` / `redoStack` — undo history (cap 50)
+
+### Annotation — `useCanvasViewportStore` (per-hub viewport, Dexie)
+- `viewMode` — `'map' | 'wall'`
+- `viewports` — per-hub zoom + pan + level state (call `rehydrateCanvasViewport(hubId)` on hub open)
+- `selection` — selected canvas nodes
+- `chartClusters` — positioned chart clusters on the Wall
+
+### Annotation — `usePreferencesStore` (per-user, idb-keyval)
+- `activeView` — workspace tab: `'dashboard' | 'frame' | 'analysis' | 'investigation' | 'improvement' | 'report'`
+- `piActiveTab` — PI panel tab: `'stats' | 'questions' | 'journal'`
+- `aiEnabled` — user AI feature toggle (Azure only)
+- `timeLens` — active time lens selection
+- `riskAxisConfig` / `budgetConfig` — improvement matrix config
+
+### Annotation — `useActiveIPStore` (per-hub per-user, localStorage)
+- `activeIPs` — map of `variscout:activeIP:{hubId}:{userId}` → `{ ipId, setAt }`
+
+### View — `useViewStore` (transient, no persist)
+- `highlightRowIndex` / `highlightedChartPoint` — bidirectional chart↔table highlight
+- `highlightedFindingId` — finding nav highlight
+- `focusedQuestionId` — Wall question focus
+- `pendingChartFocus` — set by CoScout `navigate_to` tool
+- `selectedPoints` / `selectionIndexMap` — Minitab-style brushing selection
+- `isDataTableOpen` — data table modal state
+- `improvementActiveView` — `'plan' | 'track'`
+
+## Hard Rules
+
+- **Selectors required**: `useProjectStore(s => s.field)` — never bare `useStore()`.
+- **No DataContext** (ADR-041).
+- **Document stores never import Dexie** (ESLint P7.2). R12 exception: `canvasViewportStore.ts` (separate Dexie DB for cross-app viewport state).
+- **Cross-store imperative calls are OK; reactive subscriptions between stores are forbidden**.
+- **Deleted stores**: `useSessionStore` + `useImprovementStore` were deleted in F4. Never re-introduce.
+
+## Full Glossary
+
+Complete field-level documentation (all 7 stores with types, persistence details, key actions, boundary notes):
+
+`docs/agent-context/store-state-glossary.md`
+
+Read that doc when you need the full type signatures, persistence mechanisms, or the rationale for a specific field.
+
+## Related
+
+- `packages/stores/CLAUDE.md` — authoritative one-page summary
+- ADR-078 — architecture alignment (same stores, gated tiers)
+- ADR-064 — `SuspectedCause` as first-class entity
+- `writing-tests` skill — Zustand test reset patterns (`beforeEach` + `getInitialState()`)
diff --git a/.husky/pre-push b/.husky/pre-push
index 877083aae..e8fb2c976 100755
--- a/.husky/pre-push
+++ b/.husky/pre-push
@@ -21,6 +21,19 @@ pnpm docs:check || { echo "pre-push: docs:check failed — run 'pnpm docs:check'
 echo "pre-push: running check-dead-links.sh…"
 bash "$ROOT/scripts/check-dead-links.sh" || { echo "pre-push: check-dead-links.sh failed — fix broken links and retry." >&2; exit 1; }
 
+# --- Auto-generated architecture doc staleness check ---
+# Regenerate, then fail if the committed version differs from fresh output.
+# Developers: run `pnpm docs:gen-arch && git add docs/05-technical/architecture-generated.md` before pushing.
+echo "pre-push: checking docs:gen-arch staleness…"
+pnpm docs:gen-arch > /dev/null 2>&1 && \
+  git diff --exit-code docs/05-technical/architecture-generated.md > /dev/null 2>&1 || {
+    echo "pre-push: docs/05-technical/architecture-generated.md is stale." >&2
+    echo "  Run: pnpm docs:gen-arch && git add docs/05-technical/architecture-generated.md && git commit --amend --no-edit" >&2
+    echo "  Or:  pnpm docs:gen-arch && git add docs/05-technical/architecture-generated.md && git commit -m 'chore: regenerate arch doc'" >&2
+    exit 1
+  }
+echo "pre-push: architecture-generated.md is up to date."
+
 # --- Non-blocking direct-to-main warning ---
 while read -r local_ref local_sha remote_ref remote_sha; do
   [ -z "$remote_ref" ] && continue
diff --git a/.prettierignore b/.prettierignore
new file mode 100644
index 000000000..373711d0c
--- /dev/null
+++ b/.prettierignore
@@ -0,0 +1,3 @@
+# Auto-generated by `pnpm docs:gen-arch` — script writes compact markdown tables.
+# Excluded from prettier so the staleness check in .husky/pre-push compares apples-to-apples.
+docs/05-technical/architecture-generated.md
diff --git a/CLAUDE.md b/CLAUDE.md
index 02be9a4ce..6a2deb0e3 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -6,6 +6,8 @@ Shared agent map: `docs/llms.txt`
 
 ## Invariants
 
+Full index with canonical homes + enforcement mechanisms: `.claude/INVARIANTS.md`.
+
 - Browser-only processing; data stays in customer's tenant (ADR-059). Deterministic stats engine is authority; CoScout (AI) adds context. No statistical roll-up across heterogeneous units — distributions, not aggregates (ADR-073).
 - 6 Zustand stores split across 3 layers per ADR-078 + F4 (3 Document, 2 Annotation, 1 View); no DataContext. Authoritative table: `packages/stores/CLAUDE.md`.
 - Package dependencies flow downward: core → hooks → ui → apps. Sub-path exports need **both** `package.json#exports` AND `tsconfig.json#paths` updated together. Tailwind v4: every app's `src/index.css` needs `@source` for shared packages — missing it silently breaks responsive utils.
@@ -22,13 +24,14 @@ Shared agent map: `docs/llms.txt`
 ## Where to look
 
 - **Product context**: `docs/OVERVIEW.md`, `docs/USER-JOURNEYS.md`, `docs/DATA-FLOW.md`.
+- **Strategic context**: `docs/01-vision/business-bible.md` §Strategic Hypotheses (H1–H10) — the strategic framing that underpins product decisions; `docs/01-vision/positioning.md` (how we talk about VariScout); `docs/01-vision/methodology.md` (pedagogical narrative, paired companion to vision spec §2).
 - **Operating model**: `docs/superpowers/specs/2026-04-27-process-learning-operating-model-design.md` — three-level methodology, Process Measurement System, response paths.
 - **Roadmap**: `docs/superpowers/specs/2026-04-27-product-method-roadmap-design.md` — H0–H4 horizons.
 - **Package-specific**: nested `CLAUDE.md` in each `packages/*/` and `apps/*/`.
 - **Agent manifest**: `docs/llms.txt` — map of priority entry points.
 - **Decisions (why)**: `docs/07-decisions/` (live) + `docs/archive/adrs/` (superseded).
 - **Designs (what)**: `docs/superpowers/specs/` (active) + `docs/archive/specs/` (delivered/historical).
-- **Claude-only rules + skills**: `.claude/rules/` = short cross-cutting non-negotiables (currently load globally). `.claude/skills/` = workflows only — VariScout editing patterns live in nested `CLAUDE.md`. Doc workflow: spec frontmatter SSOT in `scripts/docs-frontmatter-schema.mjs`; new docs need ≥1 inbound link.
+- **Invariants + skills**: `.claude/invariants/` = short cross-cutting non-negotiables (currently load globally). See `.claude/INVARIANTS.md` for the synthesizing index. `.claude/skills/` = workflows only — VariScout editing patterns live in nested `CLAUDE.md`. Doc workflow: spec frontmatter SSOT in `scripts/docs-frontmatter-schema.mjs`; new docs need ≥1 inbound link.
 
 ## Workflow
 
diff --git a/docs/01-vision/business-bible.md b/docs/01-vision/business-bible.md
index 11775ca5a..6f9065c5e 100644
--- a/docs/01-vision/business-bible.md
+++ b/docs/01-vision/business-bible.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: Business Bible
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [strategy, hypotheses, value-levers, flywheel, personas, pricing]
 ---
 
diff --git a/docs/01-vision/constitution.md b/docs/01-vision/constitution.md
index 749ff8d46..338b176d8 100644
--- a/docs/01-vision/constitution.md
+++ b/docs/01-vision/constitution.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'VariScout Constitution'
-audience: [developer, analyst]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [philosophy, methodology, architecture, journey]
 ---
 
diff --git a/docs/01-vision/coscout-ax-design.md b/docs/01-vision/coscout-ax-design.md
new file mode 100644
index 000000000..882024cce
--- /dev/null
+++ b/docs/01-vision/coscout-ax-design.md
@@ -0,0 +1,175 @@
+---
+title: 'CoScout AX Design (canonical)'
+purpose: design
+tier: living
+audience: [human, agent]
+topic: [ax, coscout, design]
+status: active
+related: [adr-060, adr-068, adr-069, coscout-prompts-invariant]
+last-verified: 2026-05-16
+---
+
+# CoScout AX Design (canonical)
+
+## Purpose
+
+Single canonical AX-design surface for CoScout — gathers persona, voice, tier-gated behaviors, knowledge architecture, safety boundaries, prompt engineering rules, and eval discipline that were previously scattered across ADRs 060/068/069 and invariant rules. Future CoScout design changes amend this doc; constituent ADRs and the prompt invariant remain the decision provenance and engineering detail.
+
+---
+
+## Persona + Voice
+
+**What CoScout IS:**
+
+- An investigation partner with process-improvement domain expertise
+- A coach that asks targeted questions and proposes actions grounded in visible data
+- A context carrier: surfaces relevant knowledge (investigation artifacts, external docs) when it is likely to unblock the analyst
+- A methodology ally: aware of the three-level model (System / Process Flow / Local Mechanism), the 5 analytical modes, and the 3 canvas response paths
+
+**What CoScout IS NOT:**
+
+- An oracle: it does not invent conclusions; the deterministic stats engine is the source of truth for numbers
+- A gatekeeper: it provides guidance, not checkpoints; analysts proceed at their own judgment
+- A verbose summarizer: responses are concise, grounded in context, actionable
+- A generic chatbot: every response is tied to the investigation state, the active chart, or the current question
+
+**Voice principle:** Minimal nudges over proactive interruptions (per `feedback_ai_proactivity`). CoScout speaks when it has grounded, useful context — not because it can.
+
+**Language invariant:** Never "root cause" — use "contribution," "suspected cause," or "mechanism" (P5 amended; ESLint enforced per `.claude/INVARIANTS.md`).
+
+---
+
+## Tier-Gated Behaviors
+
+CoScout behaviors differ between the PWA (free tier) and Azure (paid tier). The gate is `isPaidTier()` from `@variscout/core/tier`.
+
+| Behavior                           | PWA (free)               | Azure (paid)                              |
+| ---------------------------------- | ------------------------ | ----------------------------------------- |
+| CoScout panel visible              | Yes (read-only coaching) | Yes (full interaction)                    |
+| AI tool calls                      | No                       | Yes (27-tool registry, phase-gated)       |
+| Voice input                        | No                       | Yes (transcription-first, ADR-071)        |
+| Investigation retrieval (Pillar 2) | No                       | Yes (JSONL → Foundry IQ)                  |
+| External KB (Pillar 3)             | No                       | Yes (SOPs, FMEAs, specs)                  |
+| `answer_question` action tool      | No                       | Yes (INVESTIGATE+ phase)                  |
+| Hub connection proposals           | No                       | Yes (Converging phase only)               |
+| Prompt caching                     | N/A                      | Yes (10x token savings, tier 1 invariant) |
+
+The tier gate is **inside each surface** (signoff, audit, alerts, action proposals) — not at surface-entry CTAs. Document authoring and structured workflow surfaces serve free-tier pedagogy; `.vrs` export is always available (per `feedback_tier_gate_inside_surface`).
+
+**Implementation reference:** `packages/core/src/tier/` + `apps/azure/CLAUDE.md`.
+
+---
+
+## Knowledge Architecture
+
+Five pillars per [ADR-060](../07-decisions/adr-060-coscout-intelligence-architecture.md). Compact summary below; engineering detail in ADR-060.
+
+| Pillar                                 | What                                                                                                                                                    | Status                          |
+| -------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------- |
+| **1 — Hot Context Quality**            | `buildAIContext()` enriched with problem statement, top-5 findings with outcomes, overdue actions, focused question context, question answer visibility | Delivered (Phases 1–2)          |
+| **2 — Investigation Retrieval**        | Investigation artifacts serialized to Blob Storage as JSONL; Foundry IQ auto-indexes for on-demand retrieval via `search_knowledge_base` tool           | Pending (Azure infra Phase 3–5) |
+| **3 — External Document KB**           | SOPs, specs, FMEAs uploaded to Blob Storage; hybrid BM25 + vector + RRF search with semantic reranking; project-level security filtering                | Pending (Azure infra Phase 3–5) |
+| **4 — Question ↔ CoScout Interaction** | `answer_question` action tool (phase-gated INVESTIGATE+); question answer documents in knowledge index; hub-aware proposal cards                        | Delivered                       |
+| **5 — Mode-Aware Question Completion** | Yamazumi + performance question generators wired; mode-aware evidence sorting; `evidenceLabel`/`validationMethod`/`questionFocus` per mode              | Delivered                       |
+
+**Search backend:** Azure AI Search + Foundry IQ agentic retrieval. `KnowledgeAdapter` interface abstracts backend with fallback chain: `FoundryIQKnowledgeAdapter` → `AISearchKnowledgeAdapter` → `BlobKnowledgeAdapter`.
+
+**Mode/level orthogonality (ADR-068 amendment 2026-05-03):** Modes answer "which analytical lens?" (capability / yamazumi / defect / performance / process-flow). Levels answer "which slice of the process?" (System / Process Flow / Local Mechanism) — inferred from Canvas zoom state, not a separate picker. The two are orthogonal: any mode at any level. Level-aware overlays read Canvas zoom via `coScout/context/` pipeline; no new user gesture required.
+
+---
+
+## Safety Boundaries
+
+Three boundaries per [ADR-069](../07-decisions/adr-069-three-boundary-numeric-safety.md). Compact summary below; engineering detail in ADR-069.
+
+```
+Raw Data → [B1: Input] → Clean Data → [Stats Engine] → [B2: Output] → Safe Results → [B3: Display] → User
+```
+
+| Boundary                     | What                                                                        | Mechanism                                                         |
+| ---------------------------- | --------------------------------------------------------------------------- | ----------------------------------------------------------------- | ------------------------------------------------------------------------ |
+| **B1 — Input Sanitization**  | `toNumericValue()` rejects NaN/Infinity at parse time                       | `isFinite() && !isNaN()` in `types.ts`                            |
+| **B2 — Stats Engine Output** | Stats functions return `number                                              | undefined`, never NaN or Infinity                                 | `safeMath.ts`: `finiteOrUndefined()`, `safeDivide()`, `computeOptimum()` |
+| **B3 — Display Formatting**  | `formatStatistic()` returns "—" for non-finite values; `.toFixed()` guarded | ESLint `no-restricted-syntax` warns on `.toFixed()` in UI/AI code |
+
+**Exception:** `andersonDarlingTest()` intentionally returns `{ statistic: Infinity, pValue: 0 }` for degenerate data (all values identical). Consumers must handle this.
+
+**CoScout-specific output safety:** REF markers in AI responses reference chart elements, never raw data values (ADR-057, customer-owned data invariant). The AI prompt never receives row-level customer data — only positional chart references.
+
+---
+
+## Prompt Engineering
+
+Canonical prompt engineering rules live in `.claude/invariants/coscout-prompts.md`. Do not restate rules here; link instead.
+
+**Key constraints (brief summary):**
+
+- Entry point is `assembleCoScoutPrompt()` — `buildCoScoutSystemPrompt()` in `legacy.ts` is deprecated (test backward-compat only)
+- Every tool in `tools/registry.ts` declares `phases` + optional `tier: 'team'`; ungated tools leak across phases/tiers
+- Tier 1 is session-invariant — content that moves between tier 1 and tier 3 breaks prompt-cache hit rate (10x token cost)
+- Never "root cause" language (ESLint-enforced across prompts and CoScout code)
+
+Full rules: [`.claude/invariants/coscout-prompts.md`](../../.claude/invariants/coscout-prompts.md)
+
+---
+
+## Tool-Calling
+
+**Registry:** `packages/core/src/ai/tools/registry.ts` — 27 tools (V1 target), each gated by `phases: PhaseEnum[]` + optional `tier: 'team'`.
+
+**Phase gates:** Tools declare which investigation phases they're available in (FRAME, SCOUT, INVESTIGATE, Converging, etc.). Ungated tools appear in all phases — a correctness and UX concern, not just a permission issue.
+
+**Tier gate:** Tools marked `tier: 'team'` are stripped from prompts by `assembleCoScoutPrompt()` when `isPaidTier()` is false.
+
+**Implementation detail:** `packages/core/src/ai/tools/` — see `packages/core/CLAUDE.md` for specifics.
+
+---
+
+## Voice Input
+
+Azure-only, transcription-first (ADR-071). CoScout receives a text transcript of the analyst's voice input; it does not receive audio. Voice scope: CoScout draft input, finding editor, finding comments. Voice does not bypass investigation structure — all outputs route through the same investigation model as keyboard input.
+
+---
+
+## Eval Discipline
+
+**Current state:** No automated CoScout eval suite exists at the time of writing (2026-05-16). The promptTierSafety regression test (`packages/core/src/ai/__tests__/promptTierSafety.test.ts`) is the only mechanized quality gate.
+
+**Follow-up:** A `run-coscout-eval` skill is planned (Play 3 skills inventory). Until shipped, CoScout quality is assessed manually via representative investigation walkthroughs with a Six Sigma practitioner (Watson testing sessions — see `docs/agent-context/onboarding-quick-start.md`).
+
+**Open gap:** Log in `docs/investigations.md` — "CoScout eval suite: design and schedule automated regression tests for prompt quality, tier enforcement, phase gating, REF marker correctness."
+
+---
+
+## Mode-Aware Methodology Coaching
+
+CoScout adapts its coaching per analytical mode. Each mode has a dedicated prompt module in `packages/core/src/ai/prompts/coScout/`. Reference documents for mode-specific journeys:
+
+| Mode                          | Journey doc                          |
+| ----------------------------- | ------------------------------------ |
+| Continuous measurement (base) | `docs/USER-JOURNEYS.md`              |
+| Capability (Cp/Cpk)           | `docs/USER-JOURNEYS-CAPABILITY.md`   |
+| Yamazumi (Lean cycle-time)    | `docs/USER-JOURNEYS-YAMAZUMI.md`     |
+| Performance (multi-channel)   | `docs/USER-JOURNEYS-PERFORMANCE.md`  |
+| Defect (events-to-rates)      | `docs/USER-JOURNEYS-DEFECT.md`       |
+| Process-flow / bottleneck     | `docs/USER-JOURNEYS-PROCESS-FLOW.md` |
+
+All six journeys were updated for the wedge pivot (2026-05-16). Each inherits from the base `USER-JOURNEYS.md` and extends with mode-specific chart suite, CoScout coaching, and Specialist workflow.
+
+---
+
+## Related Artifacts
+
+- [ADR-060: CoScout Intelligence Architecture](../07-decisions/adr-060-coscout-intelligence-architecture.md) — five-pillar knowledge architecture, detail
+- [ADR-068: CoScout Cognitive Redesign](../07-decisions/adr-068-coscout-cognitive-redesign.md) — modular prompts, tier model, phase-adaptive assembly, mode/level orthogonality
+- [ADR-069: Three-Boundary Numeric Safety](../07-decisions/adr-069-three-boundary-numeric-safety.md) — B1/B2/B3 safety architecture, safeMath.ts
+- [ADR-071: CoScout Voice Input](../07-decisions/adr-071-coscout-voice-input.md) — transcription-first, Azure-only, scope
+- [ADR-057: CoScout Visual Grounding](../07-decisions/adr-057-coscout-visual-grounding.md) — REF markers, customer data boundary
+- `.claude/invariants/coscout-prompts.md` — prompt engineering non-negotiables (canonical; link from here, don't restate)
+- `packages/core/CLAUDE.md` — implementation detail: tier model, mode coaching modules, context builder
+
+---
+
+## Amendments
+
+_(Empty at V1 — 2026-05-16. Future amendments append here with date + source.)_
diff --git a/docs/01-vision/eda-mental-model.md b/docs/01-vision/eda-mental-model.md
index c9a9eb3db..7db718261 100644
--- a/docs/01-vision/eda-mental-model.md
+++ b/docs/01-vision/eda-mental-model.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'EDA Mental Model — Question-Driven Analysis'
-audience: [analyst, engineer]
+audience: human
 category: methodology
-status: stable
+status: active
 related: [methodology, four-lenses, progressive-stratification, factor-intelligence, investigation]
 ---
 
diff --git a/docs/01-vision/evaluations/bottleneck-flow-map.md b/docs/01-vision/evaluations/bottleneck-flow-map.md
index 777976c3d..77d947f0b 100644
--- a/docs/01-vision/evaluations/bottleneck-flow-map.md
+++ b/docs/01-vision/evaluations/bottleneck-flow-map.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Investigation Flow Map: Bottleneck Detection'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Investigation Flow Map: Bottleneck Detection
diff --git a/docs/01-vision/evaluations/competitive/edascout-benchmark.md b/docs/01-vision/evaluations/competitive/edascout-benchmark.md
index b121ae01f..fa642afff 100644
--- a/docs/01-vision/evaluations/competitive/edascout-benchmark.md
+++ b/docs/01-vision/evaluations/competitive/edascout-benchmark.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'EDAScout Competitive Benchmark'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # EDAScout Competitive Benchmark
diff --git a/docs/01-vision/evaluations/competitive/jmp-benchmark.md b/docs/01-vision/evaluations/competitive/jmp-benchmark.md
index 45bcd7afc..58b0c702e 100644
--- a/docs/01-vision/evaluations/competitive/jmp-benchmark.md
+++ b/docs/01-vision/evaluations/competitive/jmp-benchmark.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'JMP Competitive Benchmark'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # JMP Competitive Benchmark
diff --git a/docs/01-vision/evaluations/competitive/minitab-benchmark.md b/docs/01-vision/evaluations/competitive/minitab-benchmark.md
index 54739e290..f64d573e9 100644
--- a/docs/01-vision/evaluations/competitive/minitab-benchmark.md
+++ b/docs/01-vision/evaluations/competitive/minitab-benchmark.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Minitab Competitive Benchmark'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Minitab Competitive Benchmark
diff --git a/docs/01-vision/evaluations/competitive/minor-competitors.md b/docs/01-vision/evaluations/competitive/minor-competitors.md
index c0d694391..f6fad068c 100644
--- a/docs/01-vision/evaluations/competitive/minor-competitors.md
+++ b/docs/01-vision/evaluations/competitive/minor-competitors.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Minor Competitor Profiles'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Minor Competitor Profiles
diff --git a/docs/01-vision/evaluations/competitive/powerbi-benchmark.md b/docs/01-vision/evaluations/competitive/powerbi-benchmark.md
index 2c8075dcb..b76e06cca 100644
--- a/docs/01-vision/evaluations/competitive/powerbi-benchmark.md
+++ b/docs/01-vision/evaluations/competitive/powerbi-benchmark.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Power BI Competitive Benchmark'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Power BI Competitive Benchmark
diff --git a/docs/01-vision/evaluations/competitive/tableau-benchmark.md b/docs/01-vision/evaluations/competitive/tableau-benchmark.md
index d8a2e1881..251ae3ec1 100644
--- a/docs/01-vision/evaluations/competitive/tableau-benchmark.md
+++ b/docs/01-vision/evaluations/competitive/tableau-benchmark.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Tableau Competitive Benchmark'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Tableau Competitive Benchmark
diff --git a/docs/01-vision/evaluations/design-brief-guided-investigation.md b/docs/01-vision/evaluations/design-brief-guided-investigation.md
index d06054afd..61892b29f 100644
--- a/docs/01-vision/evaluations/design-brief-guided-investigation.md
+++ b/docs/01-vision/evaluations/design-brief-guided-investigation.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Design Brief: Guided Investigation'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Design Brief: Guided Investigation
diff --git a/docs/01-vision/evaluations/hospital-ward-flow-map.md b/docs/01-vision/evaluations/hospital-ward-flow-map.md
index 4ebbe1e88..658887120 100644
--- a/docs/01-vision/evaluations/hospital-ward-flow-map.md
+++ b/docs/01-vision/evaluations/hospital-ward-flow-map.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Investigation Flow Map: Hospital Ward (Aggregation Trap)'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Investigation Flow Map: Hospital Ward (Aggregation Trap)
diff --git a/docs/01-vision/evaluations/index.md b/docs/01-vision/evaluations/index.md
index bed5983b5..80a9e3f7e 100644
--- a/docs/01-vision/evaluations/index.md
+++ b/docs/01-vision/evaluations/index.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Evaluations: Progressive Stratification Tensions & Patterns'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Evaluations: Progressive Stratification Tensions & Patterns
diff --git a/docs/01-vision/evaluations/investigation-flow-map.md b/docs/01-vision/evaluations/investigation-flow-map.md
index 24871e107..8f5f6b1ff 100644
--- a/docs/01-vision/evaluations/investigation-flow-map.md
+++ b/docs/01-vision/evaluations/investigation-flow-map.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Investigation Flow Map: Pizza Delivery'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Investigation Flow Map: Pizza Delivery
diff --git a/docs/01-vision/evaluations/tensions/discoverability.md b/docs/01-vision/evaluations/tensions/discoverability.md
index b49356fe4..d8a272309 100644
--- a/docs/01-vision/evaluations/tensions/discoverability.md
+++ b/docs/01-vision/evaluations/tensions/discoverability.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Discoverability'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Discoverability
diff --git a/docs/01-vision/evaluations/tensions/factor-ordering.md b/docs/01-vision/evaluations/tensions/factor-ordering.md
index fe873a18b..f80320431 100644
--- a/docs/01-vision/evaluations/tensions/factor-ordering.md
+++ b/docs/01-vision/evaluations/tensions/factor-ordering.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Factor Ordering'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Factor Ordering
diff --git a/docs/01-vision/evaluations/tensions/hierarchy-assumption.md b/docs/01-vision/evaluations/tensions/hierarchy-assumption.md
index 640fad48e..bb58dcaec 100644
--- a/docs/01-vision/evaluations/tensions/hierarchy-assumption.md
+++ b/docs/01-vision/evaluations/tensions/hierarchy-assumption.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Hierarchy Assumption'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Hierarchy Assumption
diff --git a/docs/01-vision/evaluations/tensions/mobile-screen-budget.md b/docs/01-vision/evaluations/tensions/mobile-screen-budget.md
index 9033f78bc..d5e29ab4f 100644
--- a/docs/01-vision/evaluations/tensions/mobile-screen-budget.md
+++ b/docs/01-vision/evaluations/tensions/mobile-screen-budget.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Mobile Screen Budget'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Mobile Screen Budget
diff --git a/docs/01-vision/evaluations/tensions/path-dependency.md b/docs/01-vision/evaluations/tensions/path-dependency.md
index f11a6f22c..fd93793cd 100644
--- a/docs/01-vision/evaluations/tensions/path-dependency.md
+++ b/docs/01-vision/evaluations/tensions/path-dependency.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Path Dependency'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # Path Dependency
diff --git a/docs/01-vision/evaluations/tensions/when-to-stop.md b/docs/01-vision/evaluations/tensions/when-to-stop.md
index cf7299cc2..1257fc6b1 100644
--- a/docs/01-vision/evaluations/tensions/when-to-stop.md
+++ b/docs/01-vision/evaluations/tensions/when-to-stop.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'When to Stop'
-audience: [business, product]
+audience: human
 category: strategy
-status: reference
+status: active
 ---
 
 # When to Stop
diff --git a/docs/01-vision/four-lenses/change.md b/docs/01-vision/four-lenses/change.md
index c176f0c2c..ccd771b64 100644
--- a/docs/01-vision/four-lenses/change.md
+++ b/docs/01-vision/four-lenses/change.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'CHANGE Lens: I-Chart'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # CHANGE Lens: I-Chart
diff --git a/docs/01-vision/four-lenses/drilldown.md b/docs/01-vision/four-lenses/drilldown.md
index d0d250f52..64c848e92 100644
--- a/docs/01-vision/four-lenses/drilldown.md
+++ b/docs/01-vision/four-lenses/drilldown.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Drill-Down: Progressive Variation Analysis'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Drill-Down: Progressive Variation Analysis
diff --git a/docs/01-vision/four-lenses/failure.md b/docs/01-vision/four-lenses/failure.md
index 832d412ec..6c1b67d64 100644
--- a/docs/01-vision/four-lenses/failure.md
+++ b/docs/01-vision/four-lenses/failure.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'FAILURE Lens: Pareto'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # FAILURE Lens: Pareto
diff --git a/docs/01-vision/four-lenses/flow.md b/docs/01-vision/four-lenses/flow.md
index 19926986d..2ad18cada 100644
--- a/docs/01-vision/four-lenses/flow.md
+++ b/docs/01-vision/four-lenses/flow.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'FLOW Lens: Boxplot'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # FLOW Lens: Boxplot
diff --git a/docs/01-vision/four-lenses/index.md b/docs/01-vision/four-lenses/index.md
index b56f6e8a1..c35ba87f6 100644
--- a/docs/01-vision/four-lenses/index.md
+++ b/docs/01-vision/four-lenses/index.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: "Watson's Four Lenses of Process Knowledge"
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Watson's Four Lenses of Process Knowledge
diff --git a/docs/01-vision/four-lenses/value.md b/docs/01-vision/four-lenses/value.md
index edbecd502..1b49c0b81 100644
--- a/docs/01-vision/four-lenses/value.md
+++ b/docs/01-vision/four-lenses/value.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'VALUE Lens: Capability Histogram'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # VALUE Lens: Capability Histogram
diff --git a/docs/01-vision/four-lenses/yamazumi-lens.md b/docs/01-vision/four-lenses/yamazumi-lens.md
index 2e6a9cc29..e9f728a2b 100644
--- a/docs/01-vision/four-lenses/yamazumi-lens.md
+++ b/docs/01-vision/four-lenses/yamazumi-lens.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Yamazumi Lens: Composition'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Yamazumi Lens: Composition
diff --git a/docs/01-vision/index.md b/docs/01-vision/index.md
index db02c3ccc..93e4b512b 100644
--- a/docs/01-vision/index.md
+++ b/docs/01-vision/index.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Vision & Methodology'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Vision & Methodology
@@ -43,6 +45,7 @@ VariScout is built on a philosophy of **EDA for process improvement** — not st
 - [Philosophy](philosophy.md) — The EDA-for-process-improvement philosophy that drives VariScout
 - [Methodology](methodology.md) — Watson's EDA methodology, Four Lenses, Investigation Diamond, Two Voices, Three Contributions
 - [Product Overview](product-overview.md) — High-level product positioning and capabilities
+- [CoScout AX Design](coscout-ax-design.md) — Canonical CoScout design: persona, tier-gating, knowledge architecture, safety, prompt engineering
 
 ---
 
diff --git a/docs/01-vision/market-analysis.md b/docs/01-vision/market-analysis.md
index 748f6be2d..63631e67b 100644
--- a/docs/01-vision/market-analysis.md
+++ b/docs/01-vision/market-analysis.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Market Analysis & TAM Estimate'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [wedge, adr-082, product-overview, positioning, pricing, feature-parity]
 ---
diff --git a/docs/01-vision/methodology.md b/docs/01-vision/methodology.md
index 7c099b4f0..d8a428714 100644
--- a/docs/01-vision/methodology.md
+++ b/docs/01-vision/methodology.md
@@ -1,13 +1,15 @@
 ---
+tier: stable
+purpose: orient
 title: 'VariScout Methodology'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # VariScout Methodology
 
-> **Canonical product vision lives at [`docs/superpowers/specs/2026-05-03-variscout-vision-design.md`](../superpowers/specs/2026-05-03-variscout-vision-design.md).** This methodology document remains as a longer narrative companion; the vision spec's §2 (Methodology spine) is the authoritative summary. Where the two diverge, the vision spec wins. Reconciliation is a follow-up edit (see vision spec §6).
+> **Paired companion to [vision spec §2 (Methodology spine)](../superpowers/specs/2026-05-03-variscout-vision-design.md).** Two complementary canonical sources: the vision spec's §2 is the concise authoritative spine (source-of-truth for **product decisions**); this document is the longer pedagogical narrative (source-of-truth for **teaching and onboarding context**). When the two diverge, vision spec §2 wins for product decisions; this doc wins for teaching context. Reconciliation is continuous via the docs-strategy-2026 Steward loop, not a one-shot edit.
 >
 > **Canonical terminology lives at [`docs/glossary.md`](../glossary.md).** This methodology narrative cross-references the glossary rather than re-defining terms. The glossary is the home for canvas vocabulary (step / sub-step / column / input / output / outcome) and for the retired-terms list (tributary / CTS / FRAME / Analysis tab / etc) per Q10 of the 2026-05-03 vision §8 resolution.
 
diff --git a/docs/01-vision/philosophy.md b/docs/01-vision/philosophy.md
index 4032d2c2c..2da379f4d 100644
--- a/docs/01-vision/philosophy.md
+++ b/docs/01-vision/philosophy.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'EDA for Process Improvement'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # EDA for Process Improvement
diff --git a/docs/01-vision/positioning.md b/docs/01-vision/positioning.md
index 4aa4733cf..fb6dd356c 100644
--- a/docs/01-vision/positioning.md
+++ b/docs/01-vision/positioning.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Positioning Bible'
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [strategy, positioning, methodology, category, messaging]
 ---
 
@@ -18,7 +20,7 @@ related: [strategy, positioning, methodology, category, messaging]
 
 How we talk about VariScout — at every level, to every audience. A living document that captures the strategic inflection of March–April 2026 and guides all future messaging, product decisions, and documentation.
 
-Companion to the [Constitution](constitution.md) (what we believe) and the [Business Bible](business-bible.md) (how we grow).
+Companion to the [Constitution](constitution.md) (what we believe) and the [Business Bible](business-bible.md) (how we grow) — see [§Strategic Hypotheses H1–H10](business-bible.md#strategic-hypotheses) for the strategic framing that underpins this positioning.
 
 ---
 
diff --git a/docs/01-vision/product-overview.md b/docs/01-vision/product-overview.md
index c3997b286..b8c2ae553 100644
--- a/docs/01-vision/product-overview.md
+++ b/docs/01-vision/product-overview.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'VariScout: Product Overview'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [wedge, adr-082, philosophy, positioning, specifications, feature-parity, journey]
 ---
diff --git a/docs/01-vision/progressive-stratification.md b/docs/01-vision/progressive-stratification.md
index 32061a1f3..810071aa8 100644
--- a/docs/01-vision/progressive-stratification.md
+++ b/docs/01-vision/progressive-stratification.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Progressive Stratification'
-audience: [analyst, developer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [drill-down, filter-chips, eta-squared, r-squared-adj, factor-intelligence]
 ---
 
diff --git a/docs/01-vision/references/turtiainen-2019-eda-mental-model.md b/docs/01-vision/references/turtiainen-2019-eda-mental-model.md
index 746a3b8c5..803a11be3 100644
--- a/docs/01-vision/references/turtiainen-2019-eda-mental-model.md
+++ b/docs/01-vision/references/turtiainen-2019-eda-mental-model.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'EDA Mental Model Reference — Turtiainen (2019)'
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [eda-mental-model, methodology, investigation]
 ---
 
diff --git a/docs/01-vision/trainer-network.md b/docs/01-vision/trainer-network.md
index 755c05853..700c2d4b8 100644
--- a/docs/01-vision/trainer-network.md
+++ b/docs/01-vision/trainer-network.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: orient
 title: Trainer Network
-audience: [business, analyst]
+audience: human
 category: reference
 status: draft
 related: [training, ecosystem, directory, growth, market-analysis, business-bible]
diff --git a/docs/01-vision/two-voices/control-limits.md b/docs/01-vision/two-voices/control-limits.md
index e62d7cceb..a6865dc09 100644
--- a/docs/01-vision/two-voices/control-limits.md
+++ b/docs/01-vision/two-voices/control-limits.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Voice of the Process: Control Limits'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Voice of the Process: Control Limits
diff --git a/docs/01-vision/two-voices/index.md b/docs/01-vision/two-voices/index.md
index 8ceb80e88..0ec5ededd 100644
--- a/docs/01-vision/two-voices/index.md
+++ b/docs/01-vision/two-voices/index.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Two Voices: Control Limits vs Specification Limits'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Two Voices: Control Limits vs Specification Limits
diff --git a/docs/01-vision/two-voices/scenarios.md b/docs/01-vision/two-voices/scenarios.md
index 8036c97df..cb520aa47 100644
--- a/docs/01-vision/two-voices/scenarios.md
+++ b/docs/01-vision/two-voices/scenarios.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: "Four Scenarios: When the Voices Align (or Don't)"
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Four Scenarios: When the Voices Align (or Don't)
diff --git a/docs/01-vision/two-voices/spec-limits.md b/docs/01-vision/two-voices/spec-limits.md
index 0224d880a..14fc8b165 100644
--- a/docs/01-vision/two-voices/spec-limits.md
+++ b/docs/01-vision/two-voices/spec-limits.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Voice of the Customer: Specification Limits'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Voice of the Customer: Specification Limits
diff --git a/docs/01-vision/two-voices/variation-types.md b/docs/01-vision/two-voices/variation-types.md
index f343280d0..604571daa 100644
--- a/docs/01-vision/two-voices/variation-types.md
+++ b/docs/01-vision/two-voices/variation-types.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Two Types of Variation, Two Types of Action'
-audience: [business, analyst]
+audience: human
 category: methodology
-status: stable
+status: active
 ---
 
 # Two Types of Variation, Two Types of Action
diff --git a/docs/02-journeys/case-based-learning.md b/docs/02-journeys/case-based-learning.md
index 74d893b19..40aa2e61d 100644
--- a/docs/02-journeys/case-based-learning.md
+++ b/docs/02-journeys/case-based-learning.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Case-Based Learning
-audience: [analyst]
+audience: human
 category: learning
-status: stable
+status: active
 related: [case-studies, teaching, three-act-structure]
 ---
 
diff --git a/docs/02-journeys/flows/azure-admin-operations.md b/docs/02-journeys/flows/azure-admin-operations.md
index 57e44c25a..efe6af8ff 100644
--- a/docs/02-journeys/flows/azure-admin-operations.md
+++ b/docs/02-journeys/flows/azure-admin-operations.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 10: Azure App — Admin Operations'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 10: Azure App — Admin Operations
diff --git a/docs/02-journeys/flows/azure-ai-setup.md b/docs/02-journeys/flows/azure-ai-setup.md
index affcc36f7..b644e207e 100644
--- a/docs/02-journeys/flows/azure-ai-setup.md
+++ b/docs/02-journeys/flows/azure-ai-setup.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 9: Azure App — AI Setup'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 9: Azure App — AI Setup
diff --git a/docs/02-journeys/flows/azure-daily-use.md b/docs/02-journeys/flows/azure-daily-use.md
index 608c69730..e6a54a6a6 100644
--- a/docs/02-journeys/flows/azure-daily-use.md
+++ b/docs/02-journeys/flows/azure-daily-use.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 7: Azure App — Daily Use'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 7: Azure App — Daily Use
diff --git a/docs/02-journeys/flows/azure-first-analysis.md b/docs/02-journeys/flows/azure-first-analysis.md
index 8e6a7a60a..06a033cbc 100644
--- a/docs/02-journeys/flows/azure-first-analysis.md
+++ b/docs/02-journeys/flows/azure-first-analysis.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 6: Azure App — First Analysis'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 6: Azure App — First Analysis
diff --git a/docs/02-journeys/flows/azure-team-collaboration.md b/docs/02-journeys/flows/azure-team-collaboration.md
index da4d2fb9a..806593c53 100644
--- a/docs/02-journeys/flows/azure-team-collaboration.md
+++ b/docs/02-journeys/flows/azure-team-collaboration.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 8: Azure App — Team Collaboration'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 8: Azure App — Team Collaboration
diff --git a/docs/02-journeys/flows/azure-teams-mobile.md b/docs/02-journeys/flows/azure-teams-mobile.md
index e990ad530..175441b92 100644
--- a/docs/02-journeys/flows/azure-teams-mobile.md
+++ b/docs/02-journeys/flows/azure-teams-mobile.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Azure Mobile Flow'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Azure Mobile Flow
diff --git a/docs/02-journeys/flows/content-youtube.md b/docs/02-journeys/flows/content-youtube.md
index 73191898c..62ce474b8 100644
--- a/docs/02-journeys/flows/content-youtube.md
+++ b/docs/02-journeys/flows/content-youtube.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 3: YouTube/Content → Website → Product'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 3: YouTube/Content → Website → Product
diff --git a/docs/02-journeys/flows/enterprise.md b/docs/02-journeys/flows/enterprise.md
index 21fb314a2..bf202d535 100644
--- a/docs/02-journeys/flows/enterprise.md
+++ b/docs/02-journeys/flows/enterprise.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 4: Enterprise Evaluator'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 4: Enterprise Evaluator
diff --git a/docs/02-journeys/flows/factor-intelligence.md b/docs/02-journeys/flows/factor-intelligence.md
index b4b4b6254..a3ffa7d42 100644
--- a/docs/02-journeys/flows/factor-intelligence.md
+++ b/docs/02-journeys/flows/factor-intelligence.md
@@ -1,9 +1,11 @@
 ---
+tier: living
+purpose: design
 title: 'Flow: Factor Intelligence'
 journey-phase: [scout, investigate]
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow: Factor Intelligence
diff --git a/docs/02-journeys/flows/first-time.md b/docs/02-journeys/flows/first-time.md
index 8ae1ff96d..357252daf 100644
--- a/docs/02-journeys/flows/first-time.md
+++ b/docs/02-journeys/flows/first-time.md
@@ -1,7 +1,9 @@
 ---
+tier: living
+purpose: design
 title: 'First-Time Explorer Flow'
 status: draft
-audience: [product, designer]
+audience: human
 category: workflow
 ---
 
diff --git a/docs/02-journeys/flows/index.md b/docs/02-journeys/flows/index.md
index f9ae1d191..df9f58944 100644
--- a/docs/02-journeys/flows/index.md
+++ b/docs/02-journeys/flows/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: User Flows
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # User Flows
diff --git a/docs/02-journeys/flows/project-reopen.md b/docs/02-journeys/flows/project-reopen.md
index 23a3594ab..413c021f8 100644
--- a/docs/02-journeys/flows/project-reopen.md
+++ b/docs/02-journeys/flows/project-reopen.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Project Reopen Flow
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [return-visitor, azure-daily-use, project-dashboard, adr-042, adr-043]
 ---
 
diff --git a/docs/02-journeys/flows/pwa-education.md b/docs/02-journeys/flows/pwa-education.md
index eefeaf540..f81916c49 100644
--- a/docs/02-journeys/flows/pwa-education.md
+++ b/docs/02-journeys/flows/pwa-education.md
@@ -1,7 +1,9 @@
 ---
+tier: living
+purpose: design
 title: 'PWA Education Flow'
 status: draft
-audience: [product, designer]
+audience: human
 category: workflow
 ---
 
diff --git a/docs/02-journeys/flows/return-visitor.md b/docs/02-journeys/flows/return-visitor.md
index 15d5e331c..5ba2cc30c 100644
--- a/docs/02-journeys/flows/return-visitor.md
+++ b/docs/02-journeys/flows/return-visitor.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 5: Return Visitor → App'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 5: Return Visitor → App
diff --git a/docs/02-journeys/flows/seo-learner.md b/docs/02-journeys/flows/seo-learner.md
index 5d81336d6..0ff8c5513 100644
--- a/docs/02-journeys/flows/seo-learner.md
+++ b/docs/02-journeys/flows/seo-learner.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 1: SEO Learner → Product'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 1: SEO Learner → Product
diff --git a/docs/02-journeys/flows/social-discovery.md b/docs/02-journeys/flows/social-discovery.md
index 0b3ab7f62..a2b8b0a43 100644
--- a/docs/02-journeys/flows/social-discovery.md
+++ b/docs/02-journeys/flows/social-discovery.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow 2: Social Discovery → Case → Product'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow 2: Social Discovery → Case → Product
diff --git a/docs/02-journeys/flows/wide-form-analysis.md b/docs/02-journeys/flows/wide-form-analysis.md
index cb23ede7d..5323839ba 100644
--- a/docs/02-journeys/flows/wide-form-analysis.md
+++ b/docs/02-journeys/flows/wide-form-analysis.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Flow: Wide-Form Data Analysis'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Flow: Wide-Form Data Analysis
diff --git a/docs/02-journeys/index.md b/docs/02-journeys/index.md
index a455842c7..07c66e681 100644
--- a/docs/02-journeys/index.md
+++ b/docs/02-journeys/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'User Journeys'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # User Journeys
diff --git a/docs/02-journeys/personas/admin-aino.md b/docs/02-journeys/personas/admin-aino.md
index 90fe2d27d..a7dae952c 100644
--- a/docs/02-journeys/personas/admin-aino.md
+++ b/docs/02-journeys/personas/admin-aino.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Admin Aino'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Admin Aino
diff --git a/docs/02-journeys/personas/analyst-alex.md b/docs/02-journeys/personas/analyst-alex.md
index 73d9190c1..cbb4ac384 100644
--- a/docs/02-journeys/personas/analyst-alex.md
+++ b/docs/02-journeys/personas/analyst-alex.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Analyst Alex'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Analyst Alex
diff --git a/docs/02-journeys/personas/curious-carlos.md b/docs/02-journeys/personas/curious-carlos.md
index 82aa2889b..556e63800 100644
--- a/docs/02-journeys/personas/curious-carlos.md
+++ b/docs/02-journeys/personas/curious-carlos.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Curious Carlos'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Curious Carlos
diff --git a/docs/02-journeys/personas/engineer-eeva.md b/docs/02-journeys/personas/engineer-eeva.md
index 2ea6bcfec..6d793940e 100644
--- a/docs/02-journeys/personas/engineer-eeva.md
+++ b/docs/02-journeys/personas/engineer-eeva.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Engineer Eeva'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Engineer Eeva
diff --git a/docs/02-journeys/personas/evaluator-erik.md b/docs/02-journeys/personas/evaluator-erik.md
index 9ca60ee57..2f7a50df5 100644
--- a/docs/02-journeys/personas/evaluator-erik.md
+++ b/docs/02-journeys/personas/evaluator-erik.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Evaluator Erik'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Evaluator Erik
diff --git a/docs/02-journeys/personas/field-fiona.md b/docs/02-journeys/personas/field-fiona.md
index 2556b1cd5..9581fc670 100644
--- a/docs/02-journeys/personas/field-fiona.md
+++ b/docs/02-journeys/personas/field-fiona.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Field Fiona'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Field Fiona
diff --git a/docs/02-journeys/personas/green-belt-gary.md b/docs/02-journeys/personas/green-belt-gary.md
index e2af0e3a9..80462f2b8 100644
--- a/docs/02-journeys/personas/green-belt-gary.md
+++ b/docs/02-journeys/personas/green-belt-gary.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Green Belt Gary'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Green Belt Gary
diff --git a/docs/02-journeys/personas/index.md b/docs/02-journeys/personas/index.md
index b34951e1b..3f57c9c37 100644
--- a/docs/02-journeys/personas/index.md
+++ b/docs/02-journeys/personas/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Personas
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Personas
diff --git a/docs/02-journeys/personas/opex-olivia.md b/docs/02-journeys/personas/opex-olivia.md
index 9cfc8ed90..72b592ad2 100644
--- a/docs/02-journeys/personas/opex-olivia.md
+++ b/docs/02-journeys/personas/opex-olivia.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'OpEx Olivia'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # OpEx Olivia
diff --git a/docs/02-journeys/personas/student-sara.md b/docs/02-journeys/personas/student-sara.md
index 3e87d0e16..5738a44ad 100644
--- a/docs/02-journeys/personas/student-sara.md
+++ b/docs/02-journeys/personas/student-sara.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Student Sara'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Student Sara
diff --git a/docs/02-journeys/personas/trainer-tina.md b/docs/02-journeys/personas/trainer-tina.md
index 3c2b0c7e8..812de0cd4 100644
--- a/docs/02-journeys/personas/trainer-tina.md
+++ b/docs/02-journeys/personas/trainer-tina.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Trainer Tina'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Trainer Tina
diff --git a/docs/02-journeys/traceability.md b/docs/02-journeys/traceability.md
index 7c63a0b04..41ea26bd1 100644
--- a/docs/02-journeys/traceability.md
+++ b/docs/02-journeys/traceability.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Journey Traceability Index
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [flows, hooks, components, packages]
 ---
 
diff --git a/docs/02-journeys/use-cases/batch-consistency.md b/docs/02-journeys/use-cases/batch-consistency.md
index 415d70318..48b0308fe 100644
--- a/docs/02-journeys/use-cases/batch-consistency.md
+++ b/docs/02-journeys/use-cases/batch-consistency.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Batch Process Consistency'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Batch Process Consistency
diff --git a/docs/02-journeys/use-cases/bottleneck-analysis.md b/docs/02-journeys/use-cases/bottleneck-analysis.md
index e48dc8ba4..df5395815 100644
--- a/docs/02-journeys/use-cases/bottleneck-analysis.md
+++ b/docs/02-journeys/use-cases/bottleneck-analysis.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Assembly Line Bottleneck Analysis'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Assembly Line Bottleneck Analysis
diff --git a/docs/02-journeys/use-cases/call-center-performance.md b/docs/02-journeys/use-cases/call-center-performance.md
index dfc766dbd..02ac87552 100644
--- a/docs/02-journeys/use-cases/call-center-performance.md
+++ b/docs/02-journeys/use-cases/call-center-performance.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Call Center Performance Analysis'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Call Center Performance Analysis
diff --git a/docs/02-journeys/use-cases/complaint-investigation.md b/docs/02-journeys/use-cases/complaint-investigation.md
index f7a3a2042..54dd68b6e 100644
--- a/docs/02-journeys/use-cases/complaint-investigation.md
+++ b/docs/02-journeys/use-cases/complaint-investigation.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Customer Complaint Investigation'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Customer Complaint Investigation
diff --git a/docs/02-journeys/use-cases/consultant-delivery.md b/docs/02-journeys/use-cases/consultant-delivery.md
index f72067050..eef0b899b 100644
--- a/docs/02-journeys/use-cases/consultant-delivery.md
+++ b/docs/02-journeys/use-cases/consultant-delivery.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'LSS Consultant Client Delivery'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # LSS Consultant Client Delivery
diff --git a/docs/02-journeys/use-cases/copq-drilldown.md b/docs/02-journeys/use-cases/copq-drilldown.md
index a36887d37..9417aa28a 100644
--- a/docs/02-journeys/use-cases/copq-drilldown.md
+++ b/docs/02-journeys/use-cases/copq-drilldown.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'COPQ Drill-Down'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # COPQ Drill-Down
diff --git a/docs/02-journeys/use-cases/index.md b/docs/02-journeys/use-cases/index.md
index 85bf3f10c..7d49ffd67 100644
--- a/docs/02-journeys/use-cases/index.md
+++ b/docs/02-journeys/use-cases/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Use Cases'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Use Cases
diff --git a/docs/02-journeys/use-cases/lead-time-variation.md b/docs/02-journeys/use-cases/lead-time-variation.md
index 961533dc6..78615dd44 100644
--- a/docs/02-journeys/use-cases/lead-time-variation.md
+++ b/docs/02-journeys/use-cases/lead-time-variation.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Lead Time Variation Analysis'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Lead Time Variation Analysis
diff --git a/docs/02-journeys/use-cases/on-time-delivery.md b/docs/02-journeys/use-cases/on-time-delivery.md
index 5877e90d2..75c04812f 100644
--- a/docs/02-journeys/use-cases/on-time-delivery.md
+++ b/docs/02-journeys/use-cases/on-time-delivery.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'On-Time Delivery Analysis'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # On-Time Delivery Analysis
diff --git a/docs/02-journeys/use-cases/patient-wait-time.md b/docs/02-journeys/use-cases/patient-wait-time.md
index 462d5e46e..6e39fafc5 100644
--- a/docs/02-journeys/use-cases/patient-wait-time.md
+++ b/docs/02-journeys/use-cases/patient-wait-time.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Patient Wait Time Variation'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Patient Wait Time Variation
diff --git a/docs/02-journeys/use-cases/pharma-oos.md b/docs/02-journeys/use-cases/pharma-oos.md
index dac13c5ad..2d0acb557 100644
--- a/docs/02-journeys/use-cases/pharma-oos.md
+++ b/docs/02-journeys/use-cases/pharma-oos.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Pharma OOS Investigation'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Pharma OOS Investigation
diff --git a/docs/02-journeys/use-cases/supplier-performance.md b/docs/02-journeys/use-cases/supplier-performance.md
index 686205d8b..26f0d5941 100644
--- a/docs/02-journeys/use-cases/supplier-performance.md
+++ b/docs/02-journeys/use-cases/supplier-performance.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Supplier Performance Comparison'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Supplier Performance Comparison
diff --git a/docs/02-journeys/use-cases/supplier-ppap.md b/docs/02-journeys/use-cases/supplier-ppap.md
index 134fae319..762474786 100644
--- a/docs/02-journeys/use-cases/supplier-ppap.md
+++ b/docs/02-journeys/use-cases/supplier-ppap.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Supplier PPAP Capability Studies'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Supplier PPAP Capability Studies
diff --git a/docs/02-journeys/use-cases/university-spc.md b/docs/02-journeys/use-cases/university-spc.md
index 8cb9624e3..fec560bee 100644
--- a/docs/02-journeys/use-cases/university-spc.md
+++ b/docs/02-journeys/use-cases/university-spc.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'University SPC Course'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # University SPC Course
diff --git a/docs/02-journeys/ux-research.md b/docs/02-journeys/ux-research.md
index 8155afbc7..8be94751b 100644
--- a/docs/02-journeys/ux-research.md
+++ b/docs/02-journeys/ux-research.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'UX Research: VariScout Lite'
-audience: [product, designer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # UX Research: VariScout Lite
diff --git a/docs/03-features/ai/visual-grounding.md b/docs/03-features/ai/visual-grounding.md
index 4ef7992e4..c88e5213f 100644
--- a/docs/03-features/ai/visual-grounding.md
+++ b/docs/03-features/ai/visual-grounding.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: CoScout Visual Grounding
-audience: [analyst, engineer]
+audience: human
 category: ai
-status: stable
+status: active
 related: [coscout, visual-grounding, ref-markers]
 ---
 
diff --git a/docs/03-features/analysis/boxplot.md b/docs/03-features/analysis/boxplot.md
index 2c4c8df03..6bfbae663 100644
--- a/docs/03-features/analysis/boxplot.md
+++ b/docs/03-features/analysis/boxplot.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Boxplot
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [anova, eta-squared, violin-mode, factor-comparison]
 ---
 
diff --git a/docs/03-features/analysis/capability.md b/docs/03-features/analysis/capability.md
index 9ec9532de..147d1b18b 100644
--- a/docs/03-features/analysis/capability.md
+++ b/docs/03-features/analysis/capability.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Capability Analysis
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [cpk, cp, normal-distribution, sigma-estimation, specification-limits]
 ---
 
diff --git a/docs/03-features/analysis/characteristic-types.md b/docs/03-features/analysis/characteristic-types.md
index 4267d0ce3..8d8393600 100644
--- a/docs/03-features/analysis/characteristic-types.md
+++ b/docs/03-features/analysis/characteristic-types.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Characteristic Type Awareness
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [continuous, attribute, measurement-type]
 ---
 
diff --git a/docs/03-features/analysis/defect-analysis.md b/docs/03-features/analysis/defect-analysis.md
index d028d29eb..f7fb49399 100644
--- a/docs/03-features/analysis/defect-analysis.md
+++ b/docs/03-features/analysis/defect-analysis.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Defect Analysis
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [defect, pareto, aggregation, failure-mode, evidence-map]
 ---
 
diff --git a/docs/03-features/analysis/i-chart.md b/docs/03-features/analysis/i-chart.md
index bd0f4d0fb..b563166ec 100644
--- a/docs/03-features/analysis/i-chart.md
+++ b/docs/03-features/analysis/i-chart.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: I-Chart (Individuals Chart)
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [control-limits, nelson-rules, stability, time-series]
 ---
 
diff --git a/docs/03-features/analysis/index.md b/docs/03-features/analysis/index.md
index 799751650..846840349 100644
--- a/docs/03-features/analysis/index.md
+++ b/docs/03-features/analysis/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Analysis Features
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 ---
 
 # Analysis Features
diff --git a/docs/03-features/analysis/multi-level-dashboard.md b/docs/03-features/analysis/multi-level-dashboard.md
index 2b2e66548..2d6bf12cf 100644
--- a/docs/03-features/analysis/multi-level-dashboard.md
+++ b/docs/03-features/analysis/multi-level-dashboard.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Multi-Level Dashboard
-audience: [analyst]
+audience: human
 category: analysis
-status: delivered
+status: active
 related: [timeline-window-investigations, capability, process-hub-capability, stats-panel]
 ---
 
diff --git a/docs/03-features/analysis/nelson-rules.md b/docs/03-features/analysis/nelson-rules.md
index 53f0b677a..33fbaecfb 100644
--- a/docs/03-features/analysis/nelson-rules.md
+++ b/docs/03-features/analysis/nelson-rules.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Nelson Rules
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [control-chart, i-chart, special-cause, runs]
 ---
 
diff --git a/docs/03-features/analysis/pareto.md b/docs/03-features/analysis/pareto.md
index dab770106..945d79a09 100644
--- a/docs/03-features/analysis/pareto.md
+++ b/docs/03-features/analysis/pareto.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Pareto Chart
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [vital-few, contribution-percentage, drill-down]
 ---
 
diff --git a/docs/03-features/analysis/performance-mode.md b/docs/03-features/analysis/performance-mode.md
index d380a697c..fa4a653c3 100644
--- a/docs/03-features/analysis/performance-mode.md
+++ b/docs/03-features/analysis/performance-mode.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Performance Mode
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [multi-channel, cpk-comparison, wide-format]
 ---
 
diff --git a/docs/03-features/analysis/probability-plot.md b/docs/03-features/analysis/probability-plot.md
index 6f80a9af1..f23512f63 100644
--- a/docs/03-features/analysis/probability-plot.md
+++ b/docs/03-features/analysis/probability-plot.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Probability Plot
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [normal-distribution, anderson-darling, qq-plot]
 ---
 
diff --git a/docs/03-features/analysis/process-hub-capability.md b/docs/03-features/analysis/process-hub-capability.md
index 075328682..09297f05a 100644
--- a/docs/03-features/analysis/process-hub-capability.md
+++ b/docs/03-features/analysis/process-hub-capability.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Process Hub Capability Tab'
-audience: [analyst, engineer, product]
+audience: human
 category: analysis
-status: stable
+status: active
 last-reviewed: 2026-04-29
 related: [production-line-glance, process-hub, capability, hub-cadence-review]
 ---
diff --git a/docs/03-features/analysis/regression-methodology.md b/docs/03-features/analysis/regression-methodology.md
index 4266bcab2..94a80902d 100644
--- a/docs/03-features/analysis/regression-methodology.md
+++ b/docs/03-features/analysis/regression-methodology.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Regression Methodology
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [best-subsets, factor-intelligence, evidence-map, what-if]
 ---
 
diff --git a/docs/03-features/analysis/staged-analysis.md b/docs/03-features/analysis/staged-analysis.md
index 4ec7f7ebe..b6ae2c9ee 100644
--- a/docs/03-features/analysis/staged-analysis.md
+++ b/docs/03-features/analysis/staged-analysis.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Staged Analysis
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [before-after, stage-column, intervention]
 ---
 
diff --git a/docs/03-features/analysis/stats-panel.md b/docs/03-features/analysis/stats-panel.md
index 66edbc4be..2bf4a8177 100644
--- a/docs/03-features/analysis/stats-panel.md
+++ b/docs/03-features/analysis/stats-panel.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Statistics Panel
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [capability, probability-plot, specs, analysis-dashboard]
 ---
 
diff --git a/docs/03-features/analysis/subgroup-capability.md b/docs/03-features/analysis/subgroup-capability.md
index 35a2403f7..f5c0f379c 100644
--- a/docs/03-features/analysis/subgroup-capability.md
+++ b/docs/03-features/analysis/subgroup-capability.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Subgroup Capability Analysis
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [cp, cpk, rational-subgrouping, i-chart]
 ---
 
diff --git a/docs/03-features/analysis/timeline-window-investigations.md b/docs/03-features/analysis/timeline-window-investigations.md
index 60e5ce364..e7261b245 100644
--- a/docs/03-features/analysis/timeline-window-investigations.md
+++ b/docs/03-features/analysis/timeline-window-investigations.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Timeline Windows in Investigations
-audience: [analyst]
+audience: human
 category: analysis
-status: delivered
+status: active
 related: [multi-level-dashboard, capability, process-hub-capability, staged-analysis]
 ---
 
diff --git a/docs/03-features/analysis/variation-decomposition.md b/docs/03-features/analysis/variation-decomposition.md
index 9bb9f7ede..e8723195a 100644
--- a/docs/03-features/analysis/variation-decomposition.md
+++ b/docs/03-features/analysis/variation-decomposition.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Variation Decomposition
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [eta-squared, r-squared-adj, anova, factor-intelligence]
 ---
 
diff --git a/docs/03-features/analysis/yamazumi.md b/docs/03-features/analysis/yamazumi.md
index e5ca9b689..41ca4b9a2 100644
--- a/docs/03-features/analysis/yamazumi.md
+++ b/docs/03-features/analysis/yamazumi.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Yamazumi Analysis
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 related: [yamazumi, lean, time-study, takt-time, waste]
 ---
 
diff --git a/docs/03-features/data/data-input.md b/docs/03-features/data/data-input.md
index 1efdd7c35..8a9062152 100644
--- a/docs/03-features/data/data-input.md
+++ b/docs/03-features/data/data-input.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Data Input
-audience: [analyst, engineer]
+audience: human
 category: data
-status: stable
+status: active
 related: [parser, csv, excel, paste, file-upload]
 ---
 
diff --git a/docs/03-features/data/index.md b/docs/03-features/data/index.md
index df2bddd38..d48aff88a 100644
--- a/docs/03-features/data/index.md
+++ b/docs/03-features/data/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Data Features
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: stable
+status: active
 ---
 
 # Data Features
diff --git a/docs/03-features/data/storage.md b/docs/03-features/data/storage.md
index dee5a24df..8281c8879 100644
--- a/docs/03-features/data/storage.md
+++ b/docs/03-features/data/storage.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Project Persistence
-audience: [analyst, engineer]
+audience: human
 category: data
-status: stable
+status: active
 related: [indexeddb, blob-storage, browser-cache, analysis-state]
 ---
 
diff --git a/docs/03-features/data/validation.md b/docs/03-features/data/validation.md
index ffb6f203b..3641ae3e2 100644
--- a/docs/03-features/data/validation.md
+++ b/docs/03-features/data/validation.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Data Validation
-audience: [analyst, engineer]
+audience: human
 category: data
-status: stable
+status: active
 related: [parser, quality-checks, data-input]
 ---
 
diff --git a/docs/03-features/index.md b/docs/03-features/index.md
index 0179b5ad1..5d9d4f384 100644
--- a/docs/03-features/index.md
+++ b/docs/03-features/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Features
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [analysis, workflows, data, navigation, learning]
 ---
 
diff --git a/docs/03-features/learning/glossary.md b/docs/03-features/learning/glossary.md
index a48cd00cb..3812accca 100644
--- a/docs/03-features/learning/glossary.md
+++ b/docs/03-features/learning/glossary.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Glossary & Knowledge Model
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: stable
+status: active
 related: [glossary, terminology, help-tooltips, knowledge-model]
 ---
 
diff --git a/docs/03-features/navigation/breadcrumbs.md b/docs/03-features/navigation/breadcrumbs.md
index 14287fcff..dbf7b03c5 100644
--- a/docs/03-features/navigation/breadcrumbs.md
+++ b/docs/03-features/navigation/breadcrumbs.md
@@ -1,7 +1,9 @@
 ---
+tier: living
+purpose: design
 title: 'Filter Breadcrumb Navigation'
 status: draft
-audience: [analyst, engineer]
+audience: human
 category: navigation
 ---
 
diff --git a/docs/03-features/navigation/drill-down.md b/docs/03-features/navigation/drill-down.md
index 2b8e69d5d..13eeb45cf 100644
--- a/docs/03-features/navigation/drill-down.md
+++ b/docs/03-features/navigation/drill-down.md
@@ -1,7 +1,9 @@
 ---
+tier: living
+purpose: design
 title: 'Drill-Down Navigation'
 status: draft
-audience: [analyst, engineer]
+audience: human
 category: navigation
 ---
 
diff --git a/docs/03-features/navigation/progressive-filtering.md b/docs/03-features/navigation/progressive-filtering.md
index ef084f088..3108c2bc9 100644
--- a/docs/03-features/navigation/progressive-filtering.md
+++ b/docs/03-features/navigation/progressive-filtering.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Progressive Filtering
-audience: [analyst, engineer]
+audience: human
 category: navigation
-status: stable
+status: active
 related: [filter-chips, progressive-stratification, eta-squared, chart-interaction]
 ---
 
diff --git a/docs/03-features/specifications.md b/docs/03-features/specifications.md
index 63d5dd60c..f03607ca1 100644
--- a/docs/03-features/specifications.md
+++ b/docs/03-features/specifications.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: VariScout Product Spec
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [wedge, adr-082, product-overview, feature-parity, journey]
 ---
diff --git a/docs/03-features/user-guide.md b/docs/03-features/user-guide.md
index 6164f9de9..08bb640ef 100644
--- a/docs/03-features/user-guide.md
+++ b/docs/03-features/user-guide.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'VariScout Lite: User Guide'
-audience: [analyst]
+audience: human
 category: reference
-status: stable
+status: active
 related: [chart-customization, specifications, axis-editing]
 ---
 
diff --git a/docs/03-features/workflows/analysis-flow.md b/docs/03-features/workflows/analysis-flow.md
index e0aa722e9..3ab726abe 100644
--- a/docs/03-features/workflows/analysis-flow.md
+++ b/docs/03-features/workflows/analysis-flow.md
@@ -1,9 +1,11 @@
 ---
+tier: living
+purpose: design
 title: Analysis Flow
 description: Complete two-thread analysis flow — variation analysis and capability analysis — through FRAME, SCOUT, INVESTIGATE, and IMPROVE phases
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [capability, subgroup-capability, progressive-stratification, investigation-diamond]
 journey-phase: [all]
 ---
diff --git a/docs/03-features/workflows/analysis-journey-map.md b/docs/03-features/workflows/analysis-journey-map.md
index e2b25d06e..9dc7e6bcb 100644
--- a/docs/03-features/workflows/analysis-journey-map.md
+++ b/docs/03-features/workflows/analysis-journey-map.md
@@ -1,10 +1,12 @@
 ---
+tier: living
+purpose: design
 title: Analysis Journey Map
 description: Visual map of the 4-phase investigation journey with Survey readiness checks, CoScout companion, and PDCA loop
 journey-phase: [all]
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Analysis Journey Map
diff --git a/docs/03-features/workflows/decision-trees.md b/docs/03-features/workflows/decision-trees.md
index 2c0b9d493..99629b8ea 100644
--- a/docs/03-features/workflows/decision-trees.md
+++ b/docs/03-features/workflows/decision-trees.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Decision Trees
-audience: [analyst]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [chart-selection, workflow-guidance]
 journey-phase: scout
 ---
diff --git a/docs/03-features/workflows/deep-dive.md b/docs/03-features/workflows/deep-dive.md
index 666477df6..db3052278 100644
--- a/docs/03-features/workflows/deep-dive.md
+++ b/docs/03-features/workflows/deep-dive.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Deep Dive Workflow
-audience: [analyst]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [investigation, drill-down, variation-decomposition]
 ---
 
diff --git a/docs/03-features/workflows/drill-down-workflow.md b/docs/03-features/workflows/drill-down-workflow.md
index 9b099cefe..a5545d056 100644
--- a/docs/03-features/workflows/drill-down-workflow.md
+++ b/docs/03-features/workflows/drill-down-workflow.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Drill-Down Analysis Workflow
-audience: [analyst]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [drill-down, filter-chips, eta-squared, r-squared-adj, factor-intelligence]
 ---
 
diff --git a/docs/03-features/workflows/four-lenses-workflow.md b/docs/03-features/workflows/four-lenses-workflow.md
index fc8d20a28..a5c1e6914 100644
--- a/docs/03-features/workflows/four-lenses-workflow.md
+++ b/docs/03-features/workflows/four-lenses-workflow.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Four Lenses Analysis Workflow
-audience: [analyst]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [four-lenses, change, failure, flow, value]
 ---
 
diff --git a/docs/03-features/workflows/improvement-prioritization.md b/docs/03-features/workflows/improvement-prioritization.md
index bbddf5717..d5fe51121 100644
--- a/docs/03-features/workflows/improvement-prioritization.md
+++ b/docs/03-features/workflows/improvement-prioritization.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Improvement Prioritization
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [improve, prioritization, risk, timeframe, cost, matrix, ideas, actions]
 ---
 
diff --git a/docs/03-features/workflows/improvement-workspace.md b/docs/03-features/workflows/improvement-workspace.md
index 7aa5a05d9..615fd1cc5 100644
--- a/docs/03-features/workflows/improvement-workspace.md
+++ b/docs/03-features/workflows/improvement-workspace.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Improvement Workspace
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [improvement, pdca, prioritization, what-if, actions, verification]
 ---
 
diff --git a/docs/03-features/workflows/index.md b/docs/03-features/workflows/index.md
index 0e8987d2e..ef889d66f 100644
--- a/docs/03-features/workflows/index.md
+++ b/docs/03-features/workflows/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Analyst Workflows
-audience: [analyst]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [four-lenses, drill-down, investigation]
 ---
 
diff --git a/docs/03-features/workflows/investigation-lifecycle-map.md b/docs/03-features/workflows/investigation-lifecycle-map.md
index 7ec9fd448..3711d34da 100644
--- a/docs/03-features/workflows/investigation-lifecycle-map.md
+++ b/docs/03-features/workflows/investigation-lifecycle-map.md
@@ -1,10 +1,12 @@
 ---
+tier: living
+purpose: design
 title: Investigation Lifecycle Map
 description: State diagrams for investigation diamond phases and Finding status lifecycle with transition triggers and CoScout behavior
 journey-phase: investigate
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Investigation Lifecycle Map
diff --git a/docs/03-features/workflows/investigation-to-action.md b/docs/03-features/workflows/investigation-to-action.md
index 7881e2d5f..808520e8a 100644
--- a/docs/03-features/workflows/investigation-to-action.md
+++ b/docs/03-features/workflows/investigation-to-action.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Investigation to Action Workflow
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [findings, what-if, hypothesis, improvement-actions]
 ---
 
diff --git a/docs/03-features/workflows/investigation-wall.md b/docs/03-features/workflows/investigation-wall.md
index 7fdc8408a..6c4fbfdc9 100644
--- a/docs/03-features/workflows/investigation-wall.md
+++ b/docs/03-features/workflows/investigation-wall.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Investigation Wall'
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 last-reviewed: 2026-04-29
 related: [investigation-spine, evidence-map, suspected-cause, question-driven-eda, hmw-brainstorm]
 ---
diff --git a/docs/03-features/workflows/knowledge-base-search.md b/docs/03-features/workflows/knowledge-base-search.md
index 7c933013b..f491a4b9b 100644
--- a/docs/03-features/workflows/knowledge-base-search.md
+++ b/docs/03-features/workflows/knowledge-base-search.md
@@ -1,9 +1,11 @@
 ---
+tier: living
+purpose: design
 title: 'Knowledge Base Search'
 journey-phase: [scout, investigate, improve]
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Knowledge Base Search
diff --git a/docs/03-features/workflows/performance-mode-workflow.md b/docs/03-features/workflows/performance-mode-workflow.md
index e2b022bff..687234bc8 100644
--- a/docs/03-features/workflows/performance-mode-workflow.md
+++ b/docs/03-features/workflows/performance-mode-workflow.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Performance Mode Workflow
-audience: [analyst]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [performance-mode, multi-channel, cpk]
 journey-phase: scout
 ---
diff --git a/docs/03-features/workflows/process-maps.md b/docs/03-features/workflows/process-maps.md
index f33433962..9d6521d7c 100644
--- a/docs/03-features/workflows/process-maps.md
+++ b/docs/03-features/workflows/process-maps.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Visual Process Maps'
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 ---
 
 # Visual Process Maps
diff --git a/docs/03-features/workflows/project-dashboard.md b/docs/03-features/workflows/project-dashboard.md
index 2c3b182a9..98eaedf41 100644
--- a/docs/03-features/workflows/project-dashboard.md
+++ b/docs/03-features/workflows/project-dashboard.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Project Dashboard'
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [investigation, findings, hypotheses, coscout, azure, navigation]
 ---
 
diff --git a/docs/03-features/workflows/question-driven-investigation.md b/docs/03-features/workflows/question-driven-investigation.md
index f254d3424..526e35690 100644
--- a/docs/03-features/workflows/question-driven-investigation.md
+++ b/docs/03-features/workflows/question-driven-investigation.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Question-Driven Investigation Flow
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [question-tree, investigation-phases, investigation-diamond, factor-intelligence]
 ---
 
diff --git a/docs/03-features/workflows/quick-check.md b/docs/03-features/workflows/quick-check.md
index 608e103cc..e917c6f19 100644
--- a/docs/03-features/workflows/quick-check.md
+++ b/docs/03-features/workflows/quick-check.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Quick Check Workflow
-audience: [analyst]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [monitoring, shift-check, i-chart]
 ---
 
diff --git a/docs/04-cases/assembly-line/index.md b/docs/04-cases/assembly-line/index.md
index 6bb852986..8dbf778d3 100644
--- a/docs/04-cases/assembly-line/index.md
+++ b/docs/04-cases/assembly-line/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Assembly Line Case - PCB Assembly'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # Assembly Line Case - PCB Assembly
diff --git a/docs/04-cases/avocado/index.md b/docs/04-cases/avocado/index.md
index 5c2006767..5875c29b4 100644
--- a/docs/04-cases/avocado/index.md
+++ b/docs/04-cases/avocado/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Avocado Coating Case'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # Avocado Coating Case
diff --git a/docs/04-cases/bottleneck/index.md b/docs/04-cases/bottleneck/index.md
index d577a1bdf..7192a1405 100644
--- a/docs/04-cases/bottleneck/index.md
+++ b/docs/04-cases/bottleneck/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Bottleneck Case - ESTIEM'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # Bottleneck Case - ESTIEM
diff --git a/docs/04-cases/coffee/index.md b/docs/04-cases/coffee/index.md
index 5bc5ef8ce..772d7cd05 100644
--- a/docs/04-cases/coffee/index.md
+++ b/docs/04-cases/coffee/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Coffee Washing Station Case'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # Coffee Washing Station Case
diff --git a/docs/04-cases/hospital-ward/index.md b/docs/04-cases/hospital-ward/index.md
index 22608fe80..fc54ebb90 100644
--- a/docs/04-cases/hospital-ward/index.md
+++ b/docs/04-cases/hospital-ward/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Hospital Ward Case - ABB/Practitioner Context'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # Hospital Ward Case - ABB/Practitioner Context
diff --git a/docs/04-cases/index.md b/docs/04-cases/index.md
index bae7cf863..09fa49569 100644
--- a/docs/04-cases/index.md
+++ b/docs/04-cases/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'VaRiScout Lite Case Studies'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # VaRiScout Lite Case Studies
diff --git a/docs/04-cases/investigation-showcase/README.md b/docs/04-cases/investigation-showcase/README.md
index ebf92681c..d712dca18 100644
--- a/docs/04-cases/investigation-showcase/README.md
+++ b/docs/04-cases/investigation-showcase/README.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Investigation Showcase: Fill Weight'
-audience: [analyst, developer]
+audience: human
 category: workflow
-status: stable
+status: active
 related: [investigation, findings, questions, suspected-cause, regression]
 ---
 
@@ -14,12 +16,12 @@ related: [investigation, findings, questions, suspected-cause, regression]
 
 A food packaging company operates 3 filling lines across 3 shifts. Customer complaints about overfilled packages triggered an investigation.
 
-| Factor | Levels | Story |
-|--------|--------|-------|
-| Line | Line 1, Line 2, Line 3 | Line 2 has worn nozzle → mean shift +2g, 2× variation |
-| Shift | Morning, Afternoon, Night | Night shift has 1.5× variation (fatigue) |
-| Material_Batch | A, B, C, D | Batch C slightly low (ruled out — negligible effect) |
-| Operator | Kim, Lee, Park | No significant effect |
+| Factor         | Levels                    | Story                                                 |
+| -------------- | ------------------------- | ----------------------------------------------------- |
+| Line           | Line 1, Line 2, Line 3    | Line 2 has worn nozzle → mean shift +2g, 2× variation |
+| Shift          | Morning, Afternoon, Night | Night shift has 1.5× variation (fatigue)              |
+| Material_Batch | A, B, C, D                | Batch C slightly low (ruled out — negligible effect)  |
+| Operator       | Kim, Lee, Park            | No significant effect                                 |
 
 **Specs:** LSL = 495g, USL = 505g, Target = 500g
 
@@ -29,24 +31,24 @@ The dataset loads at a **mid-investigation** state:
 
 ### Questions (6)
 
-| Question | Status | Factor | Evidence |
-|----------|--------|--------|----------|
-| Does the filling line affect fill weight? | Answered | Line | η² = 0.25, R²adj = 0.23 |
-| Is Line 2's mean shift caused by nozzle wear? | Investigating | Line (gemba) | Pending inspection |
-| Does shift affect fill weight variation? | Investigating | Shift | η² = 0.08 |
-| Is night shift variation due to fatigue or maintenance? | Open | Shift (gemba) | — |
-| Does material batch affect fill weight? | Ruled out | Material_Batch | η² = 0.02 |
-| Does operator affect fill weight? | Ruled out | Operator | η² = 0.005 |
+| Question                                                | Status        | Factor         | Evidence                |
+| ------------------------------------------------------- | ------------- | -------------- | ----------------------- |
+| Does the filling line affect fill weight?               | Answered      | Line           | η² = 0.25, R²adj = 0.23 |
+| Is Line 2's mean shift caused by nozzle wear?           | Investigating | Line (gemba)   | Pending inspection      |
+| Does shift affect fill weight variation?                | Investigating | Shift          | η² = 0.08               |
+| Is night shift variation due to fatigue or maintenance? | Open          | Shift (gemba)  | —                       |
+| Does material batch affect fill weight?                 | Ruled out     | Material_Batch | η² = 0.02               |
+| Does operator affect fill weight?                       | Ruled out     | Operator       | η² = 0.005              |
 
 ### Findings (5)
 
-| Finding | Status | Key Insight |
-|---------|--------|-------------|
-| Line 2 runs consistently high (~502g) | Analyzed (key-driver) | Has What-If projection |
-| Night shift has wider spread | Investigating | Linked to shift question |
-| Batch C slightly low but within spec | Observed | Inconclusive |
-| Line 2 + Night shift worst Cpk | Investigating | Interaction effect |
-| Line 1 Morning = best-in-class | Analyzed (benchmark) | Cpk ~1.65 |
+| Finding                               | Status                | Key Insight              |
+| ------------------------------------- | --------------------- | ------------------------ |
+| Line 2 runs consistently high (~502g) | Analyzed (key-driver) | Has What-If projection   |
+| Night shift has wider spread          | Investigating         | Linked to shift question |
+| Batch C slightly low but within spec  | Observed              | Inconclusive             |
+| Line 2 + Night shift worst Cpk        | Investigating         | Interaction effect       |
+| Line 1 Morning = best-in-class        | Analyzed (benchmark)  | Cpk ~1.65                |
 
 ### Suspected Cause Hub (1)
 
@@ -79,10 +81,10 @@ Observation, Fill_Weight_g, Line, Shift, Material_Batch, Operator
 
 ## Platform Availability
 
-| Feature | PWA | Azure |
-|---------|-----|-------|
-| Data + charts | ✓ | ✓ |
-| Questions (PI Panel) | ✓ | ✓ |
-| Findings | ✓ | ✓ |
-| Suspected cause hubs | — | ✓ |
-| Improvement ideas | — | ✓ |
+| Feature              | PWA | Azure |
+| -------------------- | --- | ----- |
+| Data + charts        | ✓   | ✓     |
+| Questions (PI Panel) | ✓   | ✓     |
+| Findings             | ✓   | ✓     |
+| Suspected cause hubs | —   | ✓     |
+| Improvement ideas    | —   | ✓     |
diff --git a/docs/04-cases/machine-utilization/index.md b/docs/04-cases/machine-utilization/index.md
index c59ef0441..fafada414 100644
--- a/docs/04-cases/machine-utilization/index.md
+++ b/docs/04-cases/machine-utilization/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Case Study: "Do We Really Need a New Machine?"'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # Case Study: "Do We Really Need a New Machine?"
diff --git a/docs/04-cases/oven-zones/index.md b/docs/04-cases/oven-zones/index.md
index 44e387d4b..89a510530 100644
--- a/docs/04-cases/oven-zones/index.md
+++ b/docs/04-cases/oven-zones/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Multi-Zone Oven Temperature Control Case Study'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # Multi-Zone Oven Temperature Control Case Study
diff --git a/docs/04-cases/packaging/index.md b/docs/04-cases/packaging/index.md
index e0e0ffffa..9484e3699 100644
--- a/docs/04-cases/packaging/index.md
+++ b/docs/04-cases/packaging/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Packaging Line Case'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: reference
+status: active
 ---
 
 # Packaging Line Case
diff --git a/docs/05-technical/architecture-generated.md b/docs/05-technical/architecture-generated.md
new file mode 100644
index 000000000..2bad8967b
--- /dev/null
+++ b/docs/05-technical/architecture-generated.md
@@ -0,0 +1,92 @@
+---
+title: 'Architecture (auto-generated)'
+purpose: system
+tier: living
+audience: agent
+topic: [architecture, generated]
+status: active
+last-verified: 2026-05-17
+---
+
+> AUTO-GENERATED by `pnpm docs:gen-arch` — do not edit by hand. Regenerate after changing `package.json#dependencies`, `tsconfig.json#paths`, `package.json#exports`, or `apps/*/src/index.css` `@source` directives.
+
+# Architecture (auto-generated)
+
+Generated: 2026-05-17. Source: `scripts/docs/gen-arch.mjs`.
+
+---
+
+## Workspace Dependency Graph
+
+Internal workspace dependencies only (edges within the `@variscout/*` namespace). External npm dependencies omitted for clarity.
+
+```mermaid
+graph LR
+  Charts["charts"]
+  Core["core"]
+  Data["data"]
+  Hooks["hooks"]
+  Stores["stores"]
+  Ui["ui"]
+  AzureApp["azure-app (app)"]
+  Docs["docs (app)"]
+  Pwa["pwa (app)"]
+  Website["website (app)"]
+
+  Charts --> Core
+  Data --> Core
+  Hooks --> Core
+  Hooks --> Data
+  Hooks --> Stores
+  Stores --> Core
+  Ui --> Charts
+  Ui --> Core
+  Ui --> Hooks
+  Ui --> Stores
+  AzureApp --> Charts
+  AzureApp --> Core
+  AzureApp --> Data
+  AzureApp --> Hooks
+  AzureApp --> Stores
+  AzureApp --> Ui
+  Pwa --> Charts
+  Pwa --> Core
+  Pwa --> Data
+  Pwa --> Hooks
+  Pwa --> Stores
+  Pwa --> Ui
+  Website --> Charts
+  Website --> Core
+  Website --> Data
+  Website --> Ui
+```
+
+---
+
+## Sub-Path Export Map
+
+Sub-path exports declared in `package.json#exports` and corresponding TypeScript path aliases in `tsconfig.json#compilerOptions.paths`. These must be updated together — adding one without the other silently breaks imports. See `.claude/INVARIANTS.md` §Sub-path exports.
+
+| Package | `package.json` exports | `tsconfig.json` paths |
+|---------|------------------------|----------------------|
+| `@variscout/charts` | `.`, `./Boxplot`, `./CapabilityHistogram`, `./ChartSourceBar`, `./IChart`, `./ParetoChart`, `./ProbabilityPlot` | _(none)_ |
+| `@variscout/core` | `.`, `./actions`, `./ai`, `./canvas`, `./capability`, `./defect`, `./evidenceMap`, `./evidenceSources`, `./export`, `./findings`, `./frame`, `./glossary`, `./i18n`, `./identity`, `./improvementProject`, `./matchSummary`, `./navigation`, `./pareto`, `./parser`, `./performance`, `./persistence`, `./processHub`, `./processMoments`, `./projectMetadata`, `./responsive`, `./signalCards`, `./stats`, `./strategy`, `./survey`, `./tier`, `./time`, `./types`, `./ui-types`, `./variation`, `./yamazumi` | _(none)_ |
+| `@variscout/data` | `.`, `./computed`, `./samples` | _(none)_ |
+| `@variscout/hooks` | `.` | _(none)_ |
+| `@variscout/stores` | `.` | _(none)_ |
+| `@variscout/ui` | `.`, `./ipDetail`, `./styles/components.css`, `./styles/report-print.css`, `./styles/theme.css` | _(none)_ |
+| `@variscout/azure-app` | _(none)_ | `@/*`, `@variscout/charts`, `@variscout/charts/*`, `@variscout/core`, `@variscout/core/*`, `@variscout/data`, `@variscout/data/*`, `@variscout/hooks`, `@variscout/hooks/*`, `@variscout/ui`, `@variscout/ui/*`, `@variscout/ui/ipDetail` |
+| `@variscout/pwa` | _(none)_ | `@variscout/charts`, `@variscout/charts/*`, `@variscout/core`, `@variscout/core/*`, `@variscout/data`, `@variscout/data/*`, `@variscout/hooks`, `@variscout/hooks/*`, `@variscout/ui`, `@variscout/ui/*`, `@variscout/ui/ipDetail` |
+
+---
+
+## Tailwind v4 @source Coverage
+
+Each app must declare `@source` directives for every shared package whose class names it uses. Missing a directive silently breaks Tailwind v4 responsive utilities. See `.claude/INVARIANTS.md` §`@source` directive.
+
+| App | `@source` directives in `src/index.css` |
+|-----|------------------------------------------|
+| `@variscout/azure-app` | `../../../packages/ui/src/**/*.tsx`<br>`../../../packages/charts/src/**/*.tsx`<br>`../../../packages/hooks/src/**/*.ts` |
+| `@variscout/docs` | _(no src/index.css — not a Tailwind app)_ |
+| `@variscout/pwa` | `../../../packages/ui/src/**/*.tsx`<br>`../../../packages/charts/src/**/*.tsx`<br>`../../../packages/hooks/src/**/*.ts` |
+| `@variscout/website` | _(no src/index.css — not a Tailwind app)_ |
diff --git a/docs/05-technical/architecture.md b/docs/05-technical/architecture.md
index 6097032ff..4e820ad83 100644
--- a/docs/05-technical/architecture.md
+++ b/docs/05-technical/architecture.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'VariScout: Architecture Overview'
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [monorepo, offline-first, pwa, azure]
 ---
 
diff --git a/docs/05-technical/architecture/ai-architecture.md b/docs/05-technical/architecture/ai-architecture.md
index 2aaa85df3..7c5241a3b 100644
--- a/docs/05-technical/architecture/ai-architecture.md
+++ b/docs/05-technical/architecture/ai-architecture.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: AI Architecture
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 scope: System architecture, packages, auth, ARM, data flow, hook composition, cost controls, testing
 related: [ai-journey-integration, ai-context-engineering, aix-design-system, knowledge-model]
 ---
diff --git a/docs/05-technical/architecture/ai-context-engineering.md b/docs/05-technical/architecture/ai-context-engineering.md
index 96d305f3a..5857ea0ec 100644
--- a/docs/05-technical/architecture/ai-context-engineering.md
+++ b/docs/05-technical/architecture/ai-context-engineering.md
@@ -1,9 +1,11 @@
 ---
+tier: living
+purpose: system
 title: 'AI Context Engineering & Pipeline Reference'
 scope: Prompt tiers, context layers, token budgets, phase-aware filtering, caching
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [ai-architecture, ai-journey-integration, aix-design-system]
 points_to:
   - ai-architecture.md (system architecture, data flow, hook composition)
diff --git a/docs/05-technical/architecture/ai-journey-integration.md b/docs/05-technical/architecture/ai-journey-integration.md
index 15bfe9c15..b7213b979 100644
--- a/docs/05-technical/architecture/ai-journey-integration.md
+++ b/docs/05-technical/architecture/ai-journey-integration.md
@@ -1,9 +1,11 @@
 ---
+tier: living
+purpose: system
 title: AI Journey Integration
 scope: Entry point — AI × Phase × Mode overview, collaborator capabilities, user-facing behavior
-audience: [analyst, engineer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [ai-architecture, ai-context-engineering, aix-design-system, journey]
 points_to:
   - ai-architecture.md (system architecture, packages, auth, data flow, cost controls)
diff --git a/docs/05-technical/architecture/aix-design-system.md b/docs/05-technical/architecture/aix-design-system.md
index 6151fb66b..e70cb0081 100644
--- a/docs/05-technical/architecture/aix-design-system.md
+++ b/docs/05-technical/architecture/aix-design-system.md
@@ -1,9 +1,11 @@
 ---
+tier: living
+purpose: system
 title: AIX Design System
 scope: Governance reference — principles, tone, confidence calibration, interaction patterns, gap tracker
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [ai-architecture, ai-context-engineering, ai-journey-integration, knowledge-base]
 points_to:
   - ai-architecture.md (system architecture, data flow, cost controls)
diff --git a/docs/05-technical/architecture/capability-target-cascade.md b/docs/05-technical/architecture/capability-target-cascade.md
index 00dcedae0..4766b85b6 100644
--- a/docs/05-technical/architecture/capability-target-cascade.md
+++ b/docs/05-technical/architecture/capability-target-cascade.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Capability Target Cascade
-audience: [developer]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [analysis-strategy, dashboard-layout, capability]
 last-reviewed: 2026-04-30
 ---
diff --git a/docs/05-technical/architecture/component-map.md b/docs/05-technical/architecture/component-map.md
index 8fb02f405..7dc4a3961 100644
--- a/docs/05-technical/architecture/component-map.md
+++ b/docs/05-technical/architecture/component-map.md
@@ -1,10 +1,12 @@
 ---
+tier: living
+purpose: system
 title: Component Map
 description: L3 component views per package — internal modules and their responsibilities
 journey-phase: [all]
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # Component Map
diff --git a/docs/05-technical/architecture/component-patterns.md b/docs/05-technical/architecture/component-patterns.md
index cb8744d4f..dd8a82759 100644
--- a/docs/05-technical/architecture/component-patterns.md
+++ b/docs/05-technical/architecture/component-patterns.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Component Patterns
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [hooks, context-injection, color-scheme]
 ---
 
diff --git a/docs/05-technical/architecture/coscout-prompt-architecture.md b/docs/05-technical/architecture/coscout-prompt-architecture.md
index 28e96568b..99ab4057f 100644
--- a/docs/05-technical/architecture/coscout-prompt-architecture.md
+++ b/docs/05-technical/architecture/coscout-prompt-architecture.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: CoScout Prompt Architecture
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [ai-context-engineering, ai-architecture, adr-060, adr-049]
 ---
 
diff --git a/docs/05-technical/architecture/dashboard-layout.md b/docs/05-technical/architecture/dashboard-layout.md
index 7652792b8..184525071 100644
--- a/docs/05-technical/architecture/dashboard-layout.md
+++ b/docs/05-technical/architecture/dashboard-layout.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Dashboard Layout Architecture
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [dashboard, layout, css-grid, responsive, charts]
 ---
 
diff --git a/docs/05-technical/architecture/data-flow.md b/docs/05-technical/architecture/data-flow.md
index 68e44ae0e..e73fe7b14 100644
--- a/docs/05-technical/architecture/data-flow.md
+++ b/docs/05-technical/architecture/data-flow.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Data Flow Architecture
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [context-api, filter-stack, stats-pipeline]
 ---
 
diff --git a/docs/05-technical/architecture/data-pipeline-map.md b/docs/05-technical/architecture/data-pipeline-map.md
index db066781d..36483c3f7 100644
--- a/docs/05-technical/architecture/data-pipeline-map.md
+++ b/docs/05-technical/architecture/data-pipeline-map.md
@@ -1,10 +1,12 @@
 ---
+tier: living
+purpose: system
 title: Data Pipeline Map
 description: End-to-end data flow from CSV upload through statistics, charts, and AI — with TypeScript interfaces at every boundary
 journey-phase: [all]
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # Data Pipeline Map
diff --git a/docs/05-technical/architecture/eu-ai-act-mapping.md b/docs/05-technical/architecture/eu-ai-act-mapping.md
index bb794f79b..e32c17f27 100644
--- a/docs/05-technical/architecture/eu-ai-act-mapping.md
+++ b/docs/05-technical/architecture/eu-ai-act-mapping.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: EU AI Act Classification & Mapping
-audience: [compliance, engineer]
+audience: human
 category: architecture
-status: stable
+status: active
 date: 2026-03-19
 related: [ai, compliance, eu-ai-act, regulation]
 ---
diff --git a/docs/05-technical/architecture/index.md b/docs/05-technical/architecture/index.md
index de433fe06..e76917e83 100644
--- a/docs/05-technical/architecture/index.md
+++ b/docs/05-technical/architecture/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Architecture Documentation
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [monorepo, offline-first, data-flow, ai-architecture]
 ---
 
diff --git a/docs/05-technical/architecture/journey-phase-screen-mapping.md b/docs/05-technical/architecture/journey-phase-screen-mapping.md
index 89eedb1fd..44286243f 100644
--- a/docs/05-technical/architecture/journey-phase-screen-mapping.md
+++ b/docs/05-technical/architecture/journey-phase-screen-mapping.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Journey Phase → Screen Mapping
-audience: [analyst, engineer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [journey-phase, mental-model-hierarchy, analysis-journey-map]
 ---
 
diff --git a/docs/05-technical/architecture/knowledge-model.md b/docs/05-technical/architecture/knowledge-model.md
index 314d1b946..56fb32d35 100644
--- a/docs/05-technical/architecture/knowledge-model.md
+++ b/docs/05-technical/architecture/knowledge-model.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Knowledge Model Architecture'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # Knowledge Model Architecture
diff --git a/docs/05-technical/architecture/mental-model-hierarchy.md b/docs/05-technical/architecture/mental-model-hierarchy.md
index 014770ff2..a89ba0b3e 100644
--- a/docs/05-technical/architecture/mental-model-hierarchy.md
+++ b/docs/05-technical/architecture/mental-model-hierarchy.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Mental Model Hierarchy
-audience: [developer, analyst]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [methodology, journey, investigation, ai, report]
 ---
 
diff --git a/docs/05-technical/architecture/monorepo.md b/docs/05-technical/architecture/monorepo.md
index 1213a66b9..8130a1eef 100644
--- a/docs/05-technical/architecture/monorepo.md
+++ b/docs/05-technical/architecture/monorepo.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Monorepo Architecture'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # Monorepo Architecture
diff --git a/docs/05-technical/architecture/offline-first.md b/docs/05-technical/architecture/offline-first.md
index 7e978432f..29eb4cdc4 100644
--- a/docs/05-technical/architecture/offline-first.md
+++ b/docs/05-technical/architecture/offline-first.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Offline-First Architecture
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [service-worker, indexeddb, cache-api]
 ---
 
diff --git a/docs/05-technical/architecture/responsible-ai-policy.md b/docs/05-technical/architecture/responsible-ai-policy.md
index 9f83c2939..9bc2e6d00 100644
--- a/docs/05-technical/architecture/responsible-ai-policy.md
+++ b/docs/05-technical/architecture/responsible-ai-policy.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Responsible AI Policy
-audience: [engineer, compliance]
+audience: human
 category: architecture
-status: stable
+status: active
 date: 2026-03-19
 related: [ai, safety, compliance, eu-ai-act, guardrails]
 ---
diff --git a/docs/05-technical/architecture/shared-packages.md b/docs/05-technical/architecture/shared-packages.md
index ffbd7ed27..8792cd6cb 100644
--- a/docs/05-technical/architecture/shared-packages.md
+++ b/docs/05-technical/architecture/shared-packages.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Shared Packages'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # Shared Packages
diff --git a/docs/05-technical/architecture/store-interactions.md b/docs/05-technical/architecture/store-interactions.md
index bfe8c5a37..df1348ca5 100644
--- a/docs/05-technical/architecture/store-interactions.md
+++ b/docs/05-technical/architecture/store-interactions.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Zustand Store Interaction Graph
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [zustand, feature-stores, adr-041, state-management]
 ---
 
diff --git a/docs/05-technical/architecture/sub-path-exports.md b/docs/05-technical/architecture/sub-path-exports.md
index c2ca35cd5..7421ae13c 100644
--- a/docs/05-technical/architecture/sub-path-exports.md
+++ b/docs/05-technical/architecture/sub-path-exports.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Sub-Path Exports
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [core-package, tree-shaking, bounded-contexts]
 ---
 
diff --git a/docs/05-technical/architecture/system-map.md b/docs/05-technical/architecture/system-map.md
index 662b67dc2..5d91d829f 100644
--- a/docs/05-technical/architecture/system-map.md
+++ b/docs/05-technical/architecture/system-map.md
@@ -1,10 +1,12 @@
 ---
+tier: living
+purpose: system
 title: System Map
 description: Visual architecture overview of VariScout's packages, apps, and external integrations
 journey-phase: [all]
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # System Map
diff --git a/docs/05-technical/architecture/tier-gating.md b/docs/05-technical/architecture/tier-gating.md
index 0751e7429..e1295f7ac 100644
--- a/docs/05-technical/architecture/tier-gating.md
+++ b/docs/05-technical/architecture/tier-gating.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Tier & Plan Gating System
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [pricing, marketplace, feature-flags]
 ---
 
diff --git a/docs/05-technical/architecture/timeline-window-architecture.md b/docs/05-technical/architecture/timeline-window-architecture.md
index 8ea5a57a1..600816b8b 100644
--- a/docs/05-technical/architecture/timeline-window-architecture.md
+++ b/docs/05-technical/architecture/timeline-window-architecture.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Timeline Window & Multi-Level Routing Architecture
-audience: [developer]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [analysis-strategy, dashboard-layout, data-pipeline-map, mental-model-hierarchy]
 ---
 
diff --git a/docs/05-technical/documentation-methodology.md b/docs/05-technical/documentation-methodology.md
index 94eeac28a..5cc1f3e8d 100644
--- a/docs/05-technical/documentation-methodology.md
+++ b/docs/05-technical/documentation-methodology.md
@@ -1,10 +1,12 @@
 ---
+tier: living
+purpose: system
 title: Documentation Methodology
 description: How VariScout's 100+ docs are organized — grounded in Diataxis, C4, Docs-as-Code, and the 4-phase analysis journey
 journey-phase: [all]
-audience: [developer, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Documentation Methodology
diff --git a/docs/05-technical/implementation/azure-testing-plan.md b/docs/05-technical/implementation/azure-testing-plan.md
index 76897aa30..7d52747f9 100644
--- a/docs/05-technical/implementation/azure-testing-plan.md
+++ b/docs/05-technical/implementation/azure-testing-plan.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Azure App Testing Plan — rdmaic Organization'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Azure App Testing Plan — rdmaic Organization
diff --git a/docs/05-technical/implementation/build-system.md b/docs/05-technical/implementation/build-system.md
index cf1f6f0c6..ce53f4edf 100644
--- a/docs/05-technical/implementation/build-system.md
+++ b/docs/05-technical/implementation/build-system.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Build System & CI/CD
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 related: [turbo, vite, github-actions, deployment]
 ---
 
diff --git a/docs/05-technical/implementation/codex-only-ruflo-migration.md b/docs/05-technical/implementation/codex-only-ruflo-migration.md
index 3870e7bc7..af64b301d 100644
--- a/docs/05-technical/implementation/codex-only-ruflo-migration.md
+++ b/docs/05-technical/implementation/codex-only-ruflo-migration.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Codex-Only Ruflo Migration Guide
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Codex-Only Ruflo Migration Guide
diff --git a/docs/05-technical/implementation/codex-ruflo-workflow.md b/docs/05-technical/implementation/codex-ruflo-workflow.md
index a686ff59e..4e3ea7877 100644
--- a/docs/05-technical/implementation/codex-ruflo-workflow.md
+++ b/docs/05-technical/implementation/codex-ruflo-workflow.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Codex + Ruflo Workflow'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Codex + Ruflo Workflow
@@ -48,7 +50,7 @@ Codex does not receive the passive Claude hook guidance, so retrieval should be
 ### During Implementation
 
 - Use repo docs and package `CLAUDE.md` files for local context.
-- Treat `.claude/rules/` and `.claude/skills/` as Claude-oriented assets. Do not add Ruflo or Claude Flow skills there.
+- Treat `.claude/invariants/` and `.claude/skills/` as Claude-oriented assets. Do not add Ruflo or Claude Flow skills there.
 - Use standard repo commands for validation: `pnpm test`, `pnpm build`, and `bash scripts/pr-ready-check.sh`.
 - If you want Claude-like startup context, rely on `AGENTS.md` plus `pnpm codex:ruflo-check`; do not assume Claude hooks or statusline behavior exists in Codex.
 
diff --git a/docs/05-technical/implementation/component-extraction-guide.md b/docs/05-technical/implementation/component-extraction-guide.md
index bc7a8c8cc..8ba8907c2 100644
--- a/docs/05-technical/implementation/component-extraction-guide.md
+++ b/docs/05-technical/implementation/component-extraction-guide.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Component Extraction Guide
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 related: [shared-packages, ui-package, colorScheme]
 ---
 
diff --git a/docs/05-technical/implementation/data-input.md b/docs/05-technical/implementation/data-input.md
index 561406645..7b2e14373 100644
--- a/docs/05-technical/implementation/data-input.md
+++ b/docs/05-technical/implementation/data-input.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Data Input System'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Data Input System
diff --git a/docs/05-technical/implementation/deployment.md b/docs/05-technical/implementation/deployment.md
index 085e69561..efc00d81c 100644
--- a/docs/05-technical/implementation/deployment.md
+++ b/docs/05-technical/implementation/deployment.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Deployment Guide'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Deployment Guide
diff --git a/docs/05-technical/implementation/disaster-recovery.md b/docs/05-technical/implementation/disaster-recovery.md
index 712423606..e28dde1ef 100644
--- a/docs/05-technical/implementation/disaster-recovery.md
+++ b/docs/05-technical/implementation/disaster-recovery.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Disaster Recovery
-audience: [developer, admin]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [deployment, azure, infrastructure]
 ---
 
diff --git a/docs/05-technical/implementation/feature-module-guide.md b/docs/05-technical/implementation/feature-module-guide.md
index 66b2e7e70..86a6af6b9 100644
--- a/docs/05-technical/implementation/feature-module-guide.md
+++ b/docs/05-technical/implementation/feature-module-guide.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Feature Module Guide
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 related: [zustand, feature-sliced-design, adr-041, adr-045]
 ---
 
diff --git a/docs/05-technical/implementation/index.md b/docs/05-technical/implementation/index.md
index be142daac..880ad3120 100644
--- a/docs/05-technical/implementation/index.md
+++ b/docs/05-technical/implementation/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Implementation
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # Implementation
diff --git a/docs/05-technical/implementation/ruflo-workflow.md b/docs/05-technical/implementation/ruflo-workflow.md
index e76d65874..619af1f87 100644
--- a/docs/05-technical/implementation/ruflo-workflow.md
+++ b/docs/05-technical/implementation/ruflo-workflow.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Ruflo Development Workflow'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Ruflo Development Workflow
diff --git a/docs/05-technical/implementation/ruflo.md b/docs/05-technical/implementation/ruflo.md
index 5dde10af8..8105deca6 100644
--- a/docs/05-technical/implementation/ruflo.md
+++ b/docs/05-technical/implementation/ruflo.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Ruflo Development Tooling'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Ruflo Development Tooling
diff --git a/docs/05-technical/implementation/security-scanning.md b/docs/05-technical/implementation/security-scanning.md
index 5b44e7294..5c3eb8178 100644
--- a/docs/05-technical/implementation/security-scanning.md
+++ b/docs/05-technical/implementation/security-scanning.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Security Scanning'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Security Scanning
diff --git a/docs/05-technical/implementation/system-limits.md b/docs/05-technical/implementation/system-limits.md
index f36265956..39febdca9 100644
--- a/docs/05-technical/implementation/system-limits.md
+++ b/docs/05-technical/implementation/system-limits.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'System Limits'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # System Limits
diff --git a/docs/05-technical/implementation/testing.md b/docs/05-technical/implementation/testing.md
index 8e4822f5e..11896c1e8 100644
--- a/docs/05-technical/implementation/testing.md
+++ b/docs/05-technical/implementation/testing.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Testing Strategy'
-audience: [developer]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Testing Strategy
@@ -931,7 +933,7 @@ Validated as correctly scoped entry point. No action needed.
 
 ## Related Documentation
 
-- `.claude/rules/testing.md` - Quick reference testing rules (in project root)
+- `.claude/invariants/testing.md` - Quick reference testing invariants (in project root)
 
 ---
 
diff --git a/docs/05-technical/index.md b/docs/05-technical/index.md
index 428987727..883963372 100644
--- a/docs/05-technical/index.md
+++ b/docs/05-technical/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Technical Documentation'
-audience: [developer, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Technical Documentation
@@ -61,17 +63,18 @@ VARISCOUT ARCHITECTURE (Browser-Only)
 
 High-level architecture overview and detailed design documents:
 
-| Document                                                  | Description                                                 |
-| --------------------------------------------------------- | ----------------------------------------------------------- |
-| [Architecture Overview](architecture.md)                  | High-level system architecture                              |
-| [Monorepo Structure](architecture/monorepo.md)            | pnpm workspaces, package boundaries                         |
-| [Offline-First](architecture/offline-first.md)            | PWA, service worker, IndexedDB                              |
-| [Shared Packages](architecture/shared-packages.md)        | Package extraction and reuse strategy                       |
-| [Data Flow](architecture/data-flow.md)                    | Data pipeline from input to visualization                   |
-| [Component Patterns](architecture/component-patterns.md)  | Hook integration, colorScheme, base patterns                |
-| [AI Architecture](architecture/ai-architecture.md)        | AI integration, Azure AI Foundry, context pipeline          |
-| [AIX Design System](architecture/aix-design-system.md)    | AI experience governance: tone, trust, interaction patterns |
-| [Documentation Methodology](documentation-methodology.md) | Diataxis, C4, Docs-as-Code, journey spine                   |
+| Document                                                   | Description                                                                                             |
+| ---------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- |
+| [Architecture (auto-generated)](architecture-generated.md) | Workspace dep graph, sub-path export map, Tailwind @source coverage — generated by `pnpm docs:gen-arch` |
+| [Architecture Overview](architecture.md)                   | High-level system architecture                                                                          |
+| [Monorepo Structure](architecture/monorepo.md)             | pnpm workspaces, package boundaries                                                                     |
+| [Offline-First](architecture/offline-first.md)             | PWA, service worker, IndexedDB                                                                          |
+| [Shared Packages](architecture/shared-packages.md)         | Package extraction and reuse strategy                                                                   |
+| [Data Flow](architecture/data-flow.md)                     | Data pipeline from input to visualization                                                               |
+| [Component Patterns](architecture/component-patterns.md)   | Hook integration, colorScheme, base patterns                                                            |
+| [AI Architecture](architecture/ai-architecture.md)         | AI integration, Azure AI Foundry, context pipeline                                                      |
+| [AIX Design System](architecture/aix-design-system.md)     | AI experience governance: tone, trust, interaction patterns                                             |
+| [Documentation Methodology](documentation-methodology.md)  | Diataxis, C4, Docs-as-Code, journey spine                                                               |
 
 ### Implementation
 
diff --git a/docs/05-technical/integrations/embed-messaging.md b/docs/05-technical/integrations/embed-messaging.md
index cb50ba408..6f38ca7aa 100644
--- a/docs/05-technical/integrations/embed-messaging.md
+++ b/docs/05-technical/integrations/embed-messaging.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'PWA Embed Messaging Protocol'
-audience: [developer, architect]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # PWA Embed Messaging Protocol
diff --git a/docs/05-technical/integrations/index.md b/docs/05-technical/integrations/index.md
index 5867ee32f..b6068776a 100644
--- a/docs/05-technical/integrations/index.md
+++ b/docs/05-technical/integrations/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: Integrations
-audience: [developer]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # Integrations
diff --git a/docs/05-technical/integrations/shared-ui.md b/docs/05-technical/integrations/shared-ui.md
index ee6cd03d9..49cb4879b 100644
--- a/docs/05-technical/integrations/shared-ui.md
+++ b/docs/05-technical/integrations/shared-ui.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Shared UI Library Strategy'
-audience: [developer, architect]
+audience: human
 category: implementation
-status: stable
+status: active
 ---
 
 # Shared UI Library Strategy
diff --git a/docs/05-technical/statistics-reference.md b/docs/05-technical/statistics-reference.md
index 03698d671..0a0372042 100644
--- a/docs/05-technical/statistics-reference.md
+++ b/docs/05-technical/statistics-reference.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: 'Statistics & Investigation Technical Reference'
-audience: [developer, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Statistics & Investigation Technical Reference
diff --git a/docs/06-design-system/charts/boxplot.md b/docs/06-design-system/charts/boxplot.md
index bf45b66ab..653a2ec6c 100644
--- a/docs/06-design-system/charts/boxplot.md
+++ b/docs/06-design-system/charts/boxplot.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Boxplot Charts'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Boxplot Charts
diff --git a/docs/06-design-system/charts/capability.md b/docs/06-design-system/charts/capability.md
index d6e4a6601..6a041d686 100644
--- a/docs/06-design-system/charts/capability.md
+++ b/docs/06-design-system/charts/capability.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Capability Histograms'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Capability Histograms
diff --git a/docs/06-design-system/charts/chart-sizing-guide.md b/docs/06-design-system/charts/chart-sizing-guide.md
index 9ab0536f7..8cf80778f 100644
--- a/docs/06-design-system/charts/chart-sizing-guide.md
+++ b/docs/06-design-system/charts/chart-sizing-guide.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Chart Sizing Guide
-audience: [developer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [charts, visx, responsive, layout]
 ---
 
diff --git a/docs/06-design-system/charts/colors.md b/docs/06-design-system/charts/colors.md
index 62cdf5982..1b6dbb533 100644
--- a/docs/06-design-system/charts/colors.md
+++ b/docs/06-design-system/charts/colors.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Chart Colors'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Chart Colors
diff --git a/docs/06-design-system/charts/hooks.md b/docs/06-design-system/charts/hooks.md
index 0f74a2ae0..8b66033d9 100644
--- a/docs/06-design-system/charts/hooks.md
+++ b/docs/06-design-system/charts/hooks.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Chart Hooks'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Chart Hooks
diff --git a/docs/06-design-system/charts/ichart.md b/docs/06-design-system/charts/ichart.md
index 751f11448..293b88579 100644
--- a/docs/06-design-system/charts/ichart.md
+++ b/docs/06-design-system/charts/ichart.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Individual Control Charts (I-Chart)'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Individual Control Charts (I-Chart)
diff --git a/docs/06-design-system/charts/index.md b/docs/06-design-system/charts/index.md
index eb9d1ac58..1b3c8f59c 100644
--- a/docs/06-design-system/charts/index.md
+++ b/docs/06-design-system/charts/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Chart Design
-audience: [analyst, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Chart Design
diff --git a/docs/06-design-system/charts/overview.md b/docs/06-design-system/charts/overview.md
index 1bdbe41e0..56cf9d8d8 100644
--- a/docs/06-design-system/charts/overview.md
+++ b/docs/06-design-system/charts/overview.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Chart Styling Overview'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Chart Styling Overview
diff --git a/docs/06-design-system/charts/pareto.md b/docs/06-design-system/charts/pareto.md
index 5c41e2b43..66d73d69e 100644
--- a/docs/06-design-system/charts/pareto.md
+++ b/docs/06-design-system/charts/pareto.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Pareto Charts'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Pareto Charts
diff --git a/docs/06-design-system/charts/performance-mode.md b/docs/06-design-system/charts/performance-mode.md
index 926b5e4c5..30d7af86c 100644
--- a/docs/06-design-system/charts/performance-mode.md
+++ b/docs/06-design-system/charts/performance-mode.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Performance Mode Charts'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Performance Mode Charts
diff --git a/docs/06-design-system/charts/probability-plot.md b/docs/06-design-system/charts/probability-plot.md
index 91d165aed..6fce9dc95 100644
--- a/docs/06-design-system/charts/probability-plot.md
+++ b/docs/06-design-system/charts/probability-plot.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Probability Plot'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Probability Plot
diff --git a/docs/06-design-system/charts/responsive.md b/docs/06-design-system/charts/responsive.md
index fc5d7671e..4cec81910 100644
--- a/docs/06-design-system/charts/responsive.md
+++ b/docs/06-design-system/charts/responsive.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Responsive Charts'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Responsive Charts
diff --git a/docs/06-design-system/charts/shared-components.md b/docs/06-design-system/charts/shared-components.md
index 73d4fefc4..a99f99e88 100644
--- a/docs/06-design-system/charts/shared-components.md
+++ b/docs/06-design-system/charts/shared-components.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Shared Chart Components'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Shared Chart Components
diff --git a/docs/06-design-system/components/ai-components.md b/docs/06-design-system/components/ai-components.md
index e73d3bd46..810f3db47 100644
--- a/docs/06-design-system/components/ai-components.md
+++ b/docs/06-design-system/components/ai-components.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'AI Components'
-audience: [designer, developer]
+audience: human
 category: components
-status: stable
+status: active
 ---
 
 # AI Components
diff --git a/docs/06-design-system/components/findings.md b/docs/06-design-system/components/findings.md
index 14a0416ca..df361badc 100644
--- a/docs/06-design-system/components/findings.md
+++ b/docs/06-design-system/components/findings.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Findings'
-audience: [designer, developer]
+audience: human
 category: components
-status: stable
+status: active
 ---
 
 # Findings
diff --git a/docs/06-design-system/components/foundational-patterns.md b/docs/06-design-system/components/foundational-patterns.md
index 266b97b50..a6771bc39 100644
--- a/docs/06-design-system/components/foundational-patterns.md
+++ b/docs/06-design-system/components/foundational-patterns.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Foundational UI Patterns'
-audience: [designer, developer]
+audience: human
 category: components
-status: stable
+status: active
 ---
 
 # Foundational UI Patterns
diff --git a/docs/06-design-system/components/help-tooltip.md b/docs/06-design-system/components/help-tooltip.md
index d0cfd8cab..a3f902761 100644
--- a/docs/06-design-system/components/help-tooltip.md
+++ b/docs/06-design-system/components/help-tooltip.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Help Tooltips
-audience: [analyst, engineer]
+audience: human
 category: components
-status: stable
+status: active
 related: [glossary, contextual-help, onboarding]
 ---
 
diff --git a/docs/06-design-system/components/index.md b/docs/06-design-system/components/index.md
index 7fd23c984..72f02bd04 100644
--- a/docs/06-design-system/components/index.md
+++ b/docs/06-design-system/components/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Components
-audience: [developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Components
diff --git a/docs/06-design-system/components/signal-branch-components.md b/docs/06-design-system/components/signal-branch-components.md
index 011c5cb2c..d02e57101 100644
--- a/docs/06-design-system/components/signal-branch-components.md
+++ b/docs/06-design-system/components/signal-branch-components.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Signal And Branch Components
-audience: [designer, developer]
+audience: human
 category: reference
 status: draft
 related: [signal-cards, mechanism-branches, investigation-workspace, coscout, accessibility]
diff --git a/docs/06-design-system/components/variation-funnel.md b/docs/06-design-system/components/variation-funnel.md
index 552aa9a2d..3fc6c9422 100644
--- a/docs/06-design-system/components/variation-funnel.md
+++ b/docs/06-design-system/components/variation-funnel.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Variation Bar (Removed — ADR-062)'
-audience: [designer, developer]
+audience: human
 category: components
-status: stable
+status: active
 ---
 
 # Variation Bar (Removed)
diff --git a/docs/06-design-system/components/what-if-simulator.md b/docs/06-design-system/components/what-if-simulator.md
index c81d18754..ba495f9eb 100644
--- a/docs/06-design-system/components/what-if-simulator.md
+++ b/docs/06-design-system/components/what-if-simulator.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'What-If Simulator'
-audience: [designer, developer]
+audience: human
 category: components
-status: stable
+status: active
 ---
 
 # What-If Simulator
diff --git a/docs/06-design-system/foundations/accessibility.md b/docs/06-design-system/foundations/accessibility.md
index 67c0810d3..722c3e4fb 100644
--- a/docs/06-design-system/foundations/accessibility.md
+++ b/docs/06-design-system/foundations/accessibility.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Accessibility Guidelines'
-audience: [designer, developer]
+audience: human
 category: patterns
-status: stable
+status: active
 ---
 
 # Accessibility Guidelines
diff --git a/docs/06-design-system/foundations/colors.md b/docs/06-design-system/foundations/colors.md
index 18527588f..f47b2e4b6 100644
--- a/docs/06-design-system/foundations/colors.md
+++ b/docs/06-design-system/foundations/colors.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Color System'
-audience: [designer, developer]
+audience: human
 category: patterns
-status: stable
+status: active
 ---
 
 # Color System
diff --git a/docs/06-design-system/foundations/index.md b/docs/06-design-system/foundations/index.md
index ce72a0c56..4dbe8f3d2 100644
--- a/docs/06-design-system/foundations/index.md
+++ b/docs/06-design-system/foundations/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Foundations
-audience: [developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Foundations
diff --git a/docs/06-design-system/foundations/spacing.md b/docs/06-design-system/foundations/spacing.md
index a463adacb..066533d86 100644
--- a/docs/06-design-system/foundations/spacing.md
+++ b/docs/06-design-system/foundations/spacing.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Spacing'
-audience: [designer, developer]
+audience: human
 category: patterns
-status: stable
+status: active
 ---
 
 # Spacing
diff --git a/docs/06-design-system/foundations/typography.md b/docs/06-design-system/foundations/typography.md
index 1953d6fef..ab0771631 100644
--- a/docs/06-design-system/foundations/typography.md
+++ b/docs/06-design-system/foundations/typography.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Typography'
-audience: [designer, developer]
+audience: human
 category: patterns
-status: stable
+status: active
 ---
 
 # Typography
diff --git a/docs/06-design-system/index.md b/docs/06-design-system/index.md
index 367a4f84c..41e757451 100644
--- a/docs/06-design-system/index.md
+++ b/docs/06-design-system/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'VariScout Design System'
-audience: [designer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # VariScout Design System
diff --git a/docs/06-design-system/patterns/analysis-dashboard.md b/docs/06-design-system/patterns/analysis-dashboard.md
index 741fe1c49..0fac86912 100644
--- a/docs/06-design-system/patterns/analysis-dashboard.md
+++ b/docs/06-design-system/patterns/analysis-dashboard.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Analysis Dashboard Pattern
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [dashboard, boxplot, pareto, probability-plot, capability]
 ---
 
diff --git a/docs/06-design-system/patterns/css-height-chain.md b/docs/06-design-system/patterns/css-height-chain.md
index adb358308..5bb55bb0b 100644
--- a/docs/06-design-system/patterns/css-height-chain.md
+++ b/docs/06-design-system/patterns/css-height-chain.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: CSS Height Chain Pattern
-audience: [developer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [layout, flexbox, css-grid, dashboard]
 ---
 
diff --git a/docs/06-design-system/patterns/dashboard-design.md b/docs/06-design-system/patterns/dashboard-design.md
index 8407ff60d..d03b83539 100644
--- a/docs/06-design-system/patterns/dashboard-design.md
+++ b/docs/06-design-system/patterns/dashboard-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Dashboard Design Principles
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [analysis-dashboard, layout, dashboard, presentation-mode]
 ---
 
diff --git a/docs/06-design-system/patterns/detection-modals.md b/docs/06-design-system/patterns/detection-modals.md
index 4cc95f785..aaa8d9165 100644
--- a/docs/06-design-system/patterns/detection-modals.md
+++ b/docs/06-design-system/patterns/detection-modals.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Detection Modals
-audience: [developer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [modal, capability, performance, yamazumi, analysis-mode]
 ---
 
diff --git a/docs/06-design-system/patterns/feedback.md b/docs/06-design-system/patterns/feedback.md
index e0ee90013..9eaba6b90 100644
--- a/docs/06-design-system/patterns/feedback.md
+++ b/docs/06-design-system/patterns/feedback.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Feedback Patterns'
-audience: [designer, developer]
+audience: human
 category: patterns
-status: stable
+status: active
 ---
 
 # Feedback Patterns
diff --git a/docs/06-design-system/patterns/index.md b/docs/06-design-system/patterns/index.md
index 36ec5848a..b72cc3127 100644
--- a/docs/06-design-system/patterns/index.md
+++ b/docs/06-design-system/patterns/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Patterns
-audience: [developer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Patterns
diff --git a/docs/06-design-system/patterns/interactions.md b/docs/06-design-system/patterns/interactions.md
index 5e7702d5f..82f1f0a45 100644
--- a/docs/06-design-system/patterns/interactions.md
+++ b/docs/06-design-system/patterns/interactions.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Interaction Patterns'
-audience: [designer, developer]
+audience: human
 category: patterns
-status: stable
+status: active
 ---
 
 # Interaction Patterns
diff --git a/docs/06-design-system/patterns/investigation-workspaces.md b/docs/06-design-system/patterns/investigation-workspaces.md
index de803eab4..a114e9fff 100644
--- a/docs/06-design-system/patterns/investigation-workspaces.md
+++ b/docs/06-design-system/patterns/investigation-workspaces.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Investigation Workspace Pattern
-audience: [designer, developer]
+audience: human
 category: reference
 status: draft
 related: [investigation-wall, question-driven-eda, accessibility, ai-independent-ux, signal-cards]
diff --git a/docs/06-design-system/patterns/layout.md b/docs/06-design-system/patterns/layout.md
index f2784f163..1f4b9b177 100644
--- a/docs/06-design-system/patterns/layout.md
+++ b/docs/06-design-system/patterns/layout.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Layout Patterns'
-audience: [designer, developer]
+audience: human
 category: patterns
-status: stable
+status: active
 ---
 
 # Layout Patterns
diff --git a/docs/06-design-system/patterns/navigation.md b/docs/06-design-system/patterns/navigation.md
index 17a67c08c..f0da0df74 100644
--- a/docs/06-design-system/patterns/navigation.md
+++ b/docs/06-design-system/patterns/navigation.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Navigation Patterns
-audience: [analyst, engineer]
+audience: human
 category: patterns
-status: stable
+status: active
 related: [navigation, view-management, filter-chips, accessibility, z-index]
 ---
 
diff --git a/docs/06-design-system/patterns/panels-and-drawers.md b/docs/06-design-system/patterns/panels-and-drawers.md
index df1ac94e6..d6f0e68af 100644
--- a/docs/06-design-system/patterns/panels-and-drawers.md
+++ b/docs/06-design-system/patterns/panels-and-drawers.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Panels and Drawers'
-audience: [designer, developer]
+audience: human
 category: patterns
-status: stable
+status: active
 ---
 
 # Panels and Drawers
diff --git a/docs/06-design-system/patterns/qde2-ui-view-contracts.md b/docs/06-design-system/patterns/qde2-ui-view-contracts.md
index 66aa2568b..1215c422d 100644
--- a/docs/06-design-system/patterns/qde2-ui-view-contracts.md
+++ b/docs/06-design-system/patterns/qde2-ui-view-contracts.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: QDE 2.0 UI View Contracts
-audience: [designer, developer, product]
+audience: human
 category: reference
 status: draft
 related:
diff --git a/docs/06-design-system/patterns/tooltip-positioning.md b/docs/06-design-system/patterns/tooltip-positioning.md
index 50344cd55..f9aabaf4a 100644
--- a/docs/06-design-system/patterns/tooltip-positioning.md
+++ b/docs/06-design-system/patterns/tooltip-positioning.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Tooltip Positioning
-audience: [developer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [tooltip, viewport, positioning, help-tooltip]
 ---
 
diff --git a/docs/07-decisions/adr-001-monorepo.md b/docs/07-decisions/adr-001-monorepo.md
index 821b88b44..bbf274e92 100644
--- a/docs/07-decisions/adr-001-monorepo.md
+++ b/docs/07-decisions/adr-001-monorepo.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-001: Monorepo with pnpm Workspaces'
 ---
 
diff --git a/docs/07-decisions/adr-002-visx-charts.md b/docs/07-decisions/adr-002-visx-charts.md
index 3fa459892..212ee0570 100644
--- a/docs/07-decisions/adr-002-visx-charts.md
+++ b/docs/07-decisions/adr-002-visx-charts.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-002: Visx for Chart Components'
 ---
 
diff --git a/docs/07-decisions/adr-003-indexeddb.md b/docs/07-decisions/adr-003-indexeddb.md
index c656743bd..f186c863b 100644
--- a/docs/07-decisions/adr-003-indexeddb.md
+++ b/docs/07-decisions/adr-003-indexeddb.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-003: IndexedDB for PWA Storage'
 ---
 
diff --git a/docs/07-decisions/adr-004-offline-first.md b/docs/07-decisions/adr-004-offline-first.md
index c3d3b2184..27f6f6563 100644
--- a/docs/07-decisions/adr-004-offline-first.md
+++ b/docs/07-decisions/adr-004-offline-first.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-004: Offline-First Architecture'
 ---
 
diff --git a/docs/07-decisions/adr-005-props-based-charts.md b/docs/07-decisions/adr-005-props-based-charts.md
index a3facb318..3b48b1e91 100644
--- a/docs/07-decisions/adr-005-props-based-charts.md
+++ b/docs/07-decisions/adr-005-props-based-charts.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-005: Props-Based Charts (vs Context)'
 ---
 
diff --git a/docs/07-decisions/adr-007-azure-marketplace-distribution.md b/docs/07-decisions/adr-007-azure-marketplace-distribution.md
index 91a1f44bd..2ea50569a 100644
--- a/docs/07-decisions/adr-007-azure-marketplace-distribution.md
+++ b/docs/07-decisions/adr-007-azure-marketplace-distribution.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-007: Azure Marketplace Distribution Strategy'
 ---
 
diff --git a/docs/07-decisions/adr-008-website-content-architecture.md b/docs/07-decisions/adr-008-website-content-architecture.md
index b5fb3adf8..98d34724f 100644
--- a/docs/07-decisions/adr-008-website-content-architecture.md
+++ b/docs/07-decisions/adr-008-website-content-architecture.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-008: Website Content Architecture'
 ---
 
diff --git a/docs/07-decisions/adr-009-boxplot-violin-mode.md b/docs/07-decisions/adr-009-boxplot-violin-mode.md
index 4a042d480..58da2834e 100644
--- a/docs/07-decisions/adr-009-boxplot-violin-mode.md
+++ b/docs/07-decisions/adr-009-boxplot-violin-mode.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-009: Boxplot Violin Mode'
 ---
 
diff --git a/docs/07-decisions/adr-010-gagerr-deferral.md b/docs/07-decisions/adr-010-gagerr-deferral.md
index ba53a994b..64a871a65 100644
--- a/docs/07-decisions/adr-010-gagerr-deferral.md
+++ b/docs/07-decisions/adr-010-gagerr-deferral.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-010: Defer Gage R&R from Azure App v1'
 ---
 
diff --git a/docs/07-decisions/adr-011-ai-development-tooling.md b/docs/07-decisions/adr-011-ai-development-tooling.md
index 6edb9392a..4503fdc47 100644
--- a/docs/07-decisions/adr-011-ai-development-tooling.md
+++ b/docs/07-decisions/adr-011-ai-development-tooling.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-011: AI Development Tooling (ruflo)'
 ---
 
diff --git a/docs/07-decisions/adr-012-pwa-browser-only.md b/docs/07-decisions/adr-012-pwa-browser-only.md
index 4db5c360a..ddbee58df 100644
--- a/docs/07-decisions/adr-012-pwa-browser-only.md
+++ b/docs/07-decisions/adr-012-pwa-browser-only.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-012: PWA Browser-Only, Zero Data Collection'
 ---
 
diff --git a/docs/07-decisions/adr-013-architecture-evaluation-ddd-swarms.md b/docs/07-decisions/adr-013-architecture-evaluation-ddd-swarms.md
index bad15ea33..ea984f944 100644
--- a/docs/07-decisions/adr-013-architecture-evaluation-ddd-swarms.md
+++ b/docs/07-decisions/adr-013-architecture-evaluation-ddd-swarms.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-013: Architecture Evaluation — DDD and AI Swarms'
 ---
 
diff --git a/docs/07-decisions/adr-014-regression-deferral.md b/docs/07-decisions/adr-014-regression-deferral.md
index 07bc65b62..48201c854 100644
--- a/docs/07-decisions/adr-014-regression-deferral.md
+++ b/docs/07-decisions/adr-014-regression-deferral.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-014: Defer Regression to Phase 2'
 ---
 
diff --git a/docs/07-decisions/adr-015-investigation-board.md b/docs/07-decisions/adr-015-investigation-board.md
index 6afcdfcd7..fecf3e122 100644
--- a/docs/07-decisions/adr-015-investigation-board.md
+++ b/docs/07-decisions/adr-015-investigation-board.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-015: Investigation Board — Finding Status & Comments'
 ---
 
diff --git a/docs/07-decisions/adr-017-fluent-design-alignment.md b/docs/07-decisions/adr-017-fluent-design-alignment.md
index d795055dc..c3fd601cf 100644
--- a/docs/07-decisions/adr-017-fluent-design-alignment.md
+++ b/docs/07-decisions/adr-017-fluent-design-alignment.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-017: Fluent 2 Design Principle Alignment'
 ---
 
diff --git a/docs/07-decisions/adr-019-ai-integration.md b/docs/07-decisions/adr-019-ai-integration.md
index ac87c674d..5a3a42a3d 100644
--- a/docs/07-decisions/adr-019-ai-integration.md
+++ b/docs/07-decisions/adr-019-ai-integration.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-019: AI Integration (Azure App)'
 ---
 
diff --git a/docs/07-decisions/adr-020-investigation-workflow.md b/docs/07-decisions/adr-020-investigation-workflow.md
index 7cc9e0dcd..3af078d08 100644
--- a/docs/07-decisions/adr-020-investigation-workflow.md
+++ b/docs/07-decisions/adr-020-investigation-workflow.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-020: Investigation Workflow Enhancement'
 ---
 
diff --git a/docs/07-decisions/adr-023-data-lifecycle.md b/docs/07-decisions/adr-023-data-lifecycle.md
index 693570cb4..aed82fef1 100644
--- a/docs/07-decisions/adr-023-data-lifecycle.md
+++ b/docs/07-decisions/adr-023-data-lifecycle.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: "ADR-023: Verification Experience — Data Lifecycle, Staged Comparison, and the 'Did It Work?' Interaction"
-audience: [analyst, engineer]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-03-16
 related: [staged-analysis, findings, investigation-workflow, ai-context]
 ---
diff --git a/docs/07-decisions/adr-025-internationalization.md b/docs/07-decisions/adr-025-internationalization.md
index e1a273797..637b3ed4c 100644
--- a/docs/07-decisions/adr-025-internationalization.md
+++ b/docs/07-decisions/adr-025-internationalization.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-025: Internationalization Architecture'
 ---
 
diff --git a/docs/07-decisions/adr-027-ai-collaborator-evolution.md b/docs/07-decisions/adr-027-ai-collaborator-evolution.md
index f63e25a70..f65c707d4 100644
--- a/docs/07-decisions/adr-027-ai-collaborator-evolution.md
+++ b/docs/07-decisions/adr-027-ai-collaborator-evolution.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-027: AI Collaborator Evolution'
 ---
 
diff --git a/docs/07-decisions/adr-028-responses-api-migration.md b/docs/07-decisions/adr-028-responses-api-migration.md
index bb44a159f..00521af39 100644
--- a/docs/07-decisions/adr-028-responses-api-migration.md
+++ b/docs/07-decisions/adr-028-responses-api-migration.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-028: Responses API Migration'
 ---
 
diff --git a/docs/07-decisions/adr-029-ai-action-tools.md b/docs/07-decisions/adr-029-ai-action-tools.md
index b9255e257..7871cb68b 100644
--- a/docs/07-decisions/adr-029-ai-action-tools.md
+++ b/docs/07-decisions/adr-029-ai-action-tools.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-029: AI Action Tools for CoScout'
 ---
 
diff --git a/docs/07-decisions/adr-031-report-export.md b/docs/07-decisions/adr-031-report-export.md
index 24b1119af..4ba59cf67 100644
--- a/docs/07-decisions/adr-031-report-export.md
+++ b/docs/07-decisions/adr-031-report-export.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-031: Report Export — Print/PDF for All Azure Plans'
-audience: [analyst, engineer]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-03-19
 related: [report-view, export, pdf, print, scouting-report]
 ---
diff --git a/docs/07-decisions/adr-032-evidence-communication.md b/docs/07-decisions/adr-032-evidence-communication.md
index d353e5e23..c6cf0f8c4 100644
--- a/docs/07-decisions/adr-032-evidence-communication.md
+++ b/docs/07-decisions/adr-032-evidence-communication.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-032: Evidence-Based Statistical Communication'
-status: accepted
+status: active
 date: 2026-03-19
 ---
 
diff --git a/docs/07-decisions/adr-033-pricing-simplification.md b/docs/07-decisions/adr-033-pricing-simplification.md
index e4f8f3f96..57074d2bc 100644
--- a/docs/07-decisions/adr-033-pricing-simplification.md
+++ b/docs/07-decisions/adr-033-pricing-simplification.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-033: Pricing Simplification'
 ---
 
diff --git a/docs/07-decisions/adr-034-yamazumi-analysis-mode.md b/docs/07-decisions/adr-034-yamazumi-analysis-mode.md
index 30409a2f6..49d264f62 100644
--- a/docs/07-decisions/adr-034-yamazumi-analysis-mode.md
+++ b/docs/07-decisions/adr-034-yamazumi-analysis-mode.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-034: Yamazumi Analysis Mode'
 ---
 
diff --git a/docs/07-decisions/adr-035-improvement-prioritization.md b/docs/07-decisions/adr-035-improvement-prioritization.md
index c1d42868a..4f3ea6484 100644
--- a/docs/07-decisions/adr-035-improvement-prioritization.md
+++ b/docs/07-decisions/adr-035-improvement-prioritization.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-035: Improvement Prioritization Model'
 ---
 
diff --git a/docs/07-decisions/adr-036-no-russian-language.md b/docs/07-decisions/adr-036-no-russian-language.md
index be2d60768..0b65fe917 100644
--- a/docs/07-decisions/adr-036-no-russian-language.md
+++ b/docs/07-decisions/adr-036-no-russian-language.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-036: No Russian Language Support'
 ---
 
@@ -36,6 +38,7 @@ English, following the standard `detectLocale()` fallback chain.
 > **VariScout does not support the Russian language.**
 
 This applies to:
+
 - UI localization (message catalogs, translated strings)
 - Browser language auto-detection
 - Number/date formatting via `Intl` APIs
diff --git a/docs/07-decisions/adr-037-reporting-workspaces.md b/docs/07-decisions/adr-037-reporting-workspaces.md
index 6ac20b558..e7b510e91 100644
--- a/docs/07-decisions/adr-037-reporting-workspaces.md
+++ b/docs/07-decisions/adr-037-reporting-workspaces.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-037: Reporting Workspaces'
-audience: [analyst, engineer]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-03-20
 related: [report-view, workspace, audience-toggle, qc-story]
 ---
diff --git a/docs/07-decisions/adr-038-subgroup-capability.md b/docs/07-decisions/adr-038-subgroup-capability.md
index bd65b50a2..122fba01f 100644
--- a/docs/07-decisions/adr-038-subgroup-capability.md
+++ b/docs/07-decisions/adr-038-subgroup-capability.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-038: Subgroup Capability Analysis (Cp/Cpk per Subgroup)'
-status: accepted
+status: active
 date: 2026-03-21
 ---
 
diff --git a/docs/07-decisions/adr-039-mobile-performance-architecture.md b/docs/07-decisions/adr-039-mobile-performance-architecture.md
index 9d8a688ec..c0749f8d8 100644
--- a/docs/07-decisions/adr-039-mobile-performance-architecture.md
+++ b/docs/07-decisions/adr-039-mobile-performance-architecture.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-039: Mobile Performance & Async Computation Architecture'
 ---
 
diff --git a/docs/07-decisions/adr-040-bicep-migration.md b/docs/07-decisions/adr-040-bicep-migration.md
index 3276b88ae..d4019034f 100644
--- a/docs/07-decisions/adr-040-bicep-migration.md
+++ b/docs/07-decisions/adr-040-bicep-migration.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-040: Migrate Infrastructure from ARM JSON to Bicep'
 ---
 
diff --git a/docs/07-decisions/adr-041-zustand-feature-stores.md b/docs/07-decisions/adr-041-zustand-feature-stores.md
index 777129b03..4a23e4c4f 100644
--- a/docs/07-decisions/adr-041-zustand-feature-stores.md
+++ b/docs/07-decisions/adr-041-zustand-feature-stores.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-041: Zustand Feature Stores'
 ---
 
diff --git a/docs/07-decisions/adr-042-project-dashboard.md b/docs/07-decisions/adr-042-project-dashboard.md
index 647cb74de..cc520b494 100644
--- a/docs/07-decisions/adr-042-project-dashboard.md
+++ b/docs/07-decisions/adr-042-project-dashboard.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-042: Project Dashboard'
 ---
 
diff --git a/docs/07-decisions/adr-044-architectural-review.md b/docs/07-decisions/adr-044-architectural-review.md
index a86e7c35d..89384e09e 100644
--- a/docs/07-decisions/adr-044-architectural-review.md
+++ b/docs/07-decisions/adr-044-architectural-review.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-044: Architectural Review (March 2026)'
-status: accepted
+status: active
 date: 2026-03-23
 ---
 
diff --git a/docs/07-decisions/adr-045-modular-architecture.md b/docs/07-decisions/adr-045-modular-architecture.md
index 8c02a85e7..4a7af097a 100644
--- a/docs/07-decisions/adr-045-modular-architecture.md
+++ b/docs/07-decisions/adr-045-modular-architecture.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-045: Modular Architecture — DDD-Lite with Feature-Sliced Design'
 ---
 
diff --git a/docs/07-decisions/adr-046-event-driven-architecture.md b/docs/07-decisions/adr-046-event-driven-architecture.md
index 07c96537f..8055c540c 100644
--- a/docs/07-decisions/adr-046-event-driven-architecture.md
+++ b/docs/07-decisions/adr-046-event-driven-architecture.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-046: Event-Driven Architecture — mitt Event Bus'
 ---
 
diff --git a/docs/07-decisions/adr-047-analysis-mode-strategy.md b/docs/07-decisions/adr-047-analysis-mode-strategy.md
index c131dfeb5..fca6678e3 100644
--- a/docs/07-decisions/adr-047-analysis-mode-strategy.md
+++ b/docs/07-decisions/adr-047-analysis-mode-strategy.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-047: Analysis Mode Strategy Pattern'
 ---
 
diff --git a/docs/07-decisions/adr-048-eslint-boundaries.md b/docs/07-decisions/adr-048-eslint-boundaries.md
index 469258043..8f9a1fb16 100644
--- a/docs/07-decisions/adr-048-eslint-boundaries.md
+++ b/docs/07-decisions/adr-048-eslint-boundaries.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-048: ESLint Boundary Enforcement'
 ---
 
diff --git a/docs/07-decisions/adr-049-coscout-context-and-memory.md b/docs/07-decisions/adr-049-coscout-context-and-memory.md
index 08319bf71..c3f0173ec 100644
--- a/docs/07-decisions/adr-049-coscout-context-and-memory.md
+++ b/docs/07-decisions/adr-049-coscout-context-and-memory.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-049: CoScout Knowledge Catalyst — Visual Context & Active Insight Capture'
 ---
 
diff --git a/docs/07-decisions/adr-050-wide-form-stack-columns.md b/docs/07-decisions/adr-050-wide-form-stack-columns.md
index d1ab6345c..f6bfcc3bd 100644
--- a/docs/07-decisions/adr-050-wide-form-stack-columns.md
+++ b/docs/07-decisions/adr-050-wide-form-stack-columns.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-050: Wide-Form Data Support — Stack Columns'
 ---
 
diff --git a/docs/07-decisions/adr-051-chart-many-categories.md b/docs/07-decisions/adr-051-chart-many-categories.md
index 86537c166..933fa7f1b 100644
--- a/docs/07-decisions/adr-051-chart-many-categories.md
+++ b/docs/07-decisions/adr-051-chart-many-categories.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-051: Chart Handling for Many Categories'
 ---
 
diff --git a/docs/07-decisions/adr-052-factor-intelligence.md b/docs/07-decisions/adr-052-factor-intelligence.md
index b65bd2c00..acbe1d676 100644
--- a/docs/07-decisions/adr-052-factor-intelligence.md
+++ b/docs/07-decisions/adr-052-factor-intelligence.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-052: Factor Intelligence — Progressive Factor Analysis'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [statistics, factor-analysis, best-subsets, strategy-pattern]
 ---
 
diff --git a/docs/07-decisions/adr-053-question-driven-investigation.md b/docs/07-decisions/adr-053-question-driven-investigation.md
index 90f3110ef..6169f73db 100644
--- a/docs/07-decisions/adr-053-question-driven-investigation.md
+++ b/docs/07-decisions/adr-053-question-driven-investigation.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-053: Question-Driven Investigation'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [investigation, eda-mental-model, factor-intelligence, question-driven-eda]
 ---
 
diff --git a/docs/07-decisions/adr-054-mode-aware-question-strategy.md b/docs/07-decisions/adr-054-mode-aware-question-strategy.md
index 8fd844bc5..451eb3c5b 100644
--- a/docs/07-decisions/adr-054-mode-aware-question-strategy.md
+++ b/docs/07-decisions/adr-054-mode-aware-question-strategy.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-054: Mode-Aware Question Strategy'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [question-driven-eda, factor-intelligence, strategy-pattern, yamazumi, capability]
 ---
 
diff --git a/docs/07-decisions/adr-055-workspace-navigation.md b/docs/07-decisions/adr-055-workspace-navigation.md
index b03a035b9..464df975a 100644
--- a/docs/07-decisions/adr-055-workspace-navigation.md
+++ b/docs/07-decisions/adr-055-workspace-navigation.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-055: Workspace-Based Navigation'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [navigation, workspaces, investigation, panels, question-driven-eda]
 ---
 
diff --git a/docs/07-decisions/adr-056-pi-panel-redesign.md b/docs/07-decisions/adr-056-pi-panel-redesign.md
index b9d94b5a1..1d805418a 100644
--- a/docs/07-decisions/adr-056-pi-panel-redesign.md
+++ b/docs/07-decisions/adr-056-pi-panel-redesign.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-056: Process Intelligence Panel Redesign'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 related: [pi-panel, questions, journal, investigation, findings, stats-panel]
 ---
 
diff --git a/docs/07-decisions/adr-057-coscout-visual-grounding.md b/docs/07-decisions/adr-057-coscout-visual-grounding.md
index 421786b05..ccd403727 100644
--- a/docs/07-decisions/adr-057-coscout-visual-grounding.md
+++ b/docs/07-decisions/adr-057-coscout-visual-grounding.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-057: CoScout Visual Grounding'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # ADR-057: CoScout Visual Grounding
diff --git a/docs/07-decisions/adr-058-deployment-lifecycle.md b/docs/07-decisions/adr-058-deployment-lifecycle.md
index 9beab52f8..bd9469a5c 100644
--- a/docs/07-decisions/adr-058-deployment-lifecycle.md
+++ b/docs/07-decisions/adr-058-deployment-lifecycle.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-058: Deployment Lifecycle & Self-Service Update System'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # ADR-058: Deployment Lifecycle & Self-Service Update System
diff --git a/docs/07-decisions/adr-059-web-first-deployment-architecture.md b/docs/07-decisions/adr-059-web-first-deployment-architecture.md
index 6df12bdbd..c6cf6e5ad 100644
--- a/docs/07-decisions/adr-059-web-first-deployment-architecture.md
+++ b/docs/07-decisions/adr-059-web-first-deployment-architecture.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-059: Web-First Deployment Architecture'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 ---
 
 # ADR-059: Web-First Deployment Architecture
diff --git a/docs/07-decisions/adr-060-coscout-intelligence-architecture.md b/docs/07-decisions/adr-060-coscout-intelligence-architecture.md
index 8f9259c8d..5d03ca24c 100644
--- a/docs/07-decisions/adr-060-coscout-intelligence-architecture.md
+++ b/docs/07-decisions/adr-060-coscout-intelligence-architecture.md
@@ -1,12 +1,17 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-060: CoScout Intelligence Architecture'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
+related: [coscout-ax-design]
 ---
 
 # ADR-060: CoScout Intelligence Architecture
 
+> See canonical AX design at [coscout-ax-design](../01-vision/coscout-ax-design.md) for the consolidated persona, tier-gating, safety, and prompt engineering overview. This ADR is the decision provenance and engineering detail for the five-pillar knowledge architecture.
+
 **Status:** Accepted
 **Date:** 2026-04-02
 **Supersedes:** ADR-022 (Knowledge Layer Architecture), ADR-026 (Knowledge Base — SharePoint-First)
diff --git a/docs/07-decisions/adr-061-hmw-brainstorm-ideation.md b/docs/07-decisions/adr-061-hmw-brainstorm-ideation.md
index c9949ba39..11f225771 100644
--- a/docs/07-decisions/adr-061-hmw-brainstorm-ideation.md
+++ b/docs/07-decisions/adr-061-hmw-brainstorm-ideation.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-061: HMW Brainstorm & Collaborative Ideation'
 ---
 
diff --git a/docs/07-decisions/adr-062-standard-anova-metrics.md b/docs/07-decisions/adr-062-standard-anova-metrics.md
index 37ba1e24c..1c4e7a795 100644
--- a/docs/07-decisions/adr-062-standard-anova-metrics.md
+++ b/docs/07-decisions/adr-062-standard-anova-metrics.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-062: Standard ANOVA Metrics & Category Total SS Removal'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: stable
+status: active
 related:
   [anova, factor-intelligence, question-driven-investigation, variation-decomposition, eta-squared]
 ---
diff --git a/docs/07-decisions/adr-063-trust-compliance-roadmap.md b/docs/07-decisions/adr-063-trust-compliance-roadmap.md
index a2ba3e226..14b863909 100644
--- a/docs/07-decisions/adr-063-trust-compliance-roadmap.md
+++ b/docs/07-decisions/adr-063-trust-compliance-roadmap.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-063: Trust & Compliance Roadmap'
-audience: [business, architect]
+audience: human
 category: compliance
-status: accepted
+status: active
 ---
 
 # ADR-063: Trust & Compliance Roadmap
diff --git a/docs/07-decisions/adr-065-evidence-map-causal-graph.md b/docs/07-decisions/adr-065-evidence-map-causal-graph.md
index 496414a02..edd0c7c1f 100644
--- a/docs/07-decisions/adr-065-evidence-map-causal-graph.md
+++ b/docs/07-decisions/adr-065-evidence-map-causal-graph.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-065: Evidence Map & Causal Graph Visualization'
 ---
 
diff --git a/docs/07-decisions/adr-066-evidence-map-investigation-center.md b/docs/07-decisions/adr-066-evidence-map-investigation-center.md
index 87367de8c..c46fcdca4 100644
--- a/docs/07-decisions/adr-066-evidence-map-investigation-center.md
+++ b/docs/07-decisions/adr-066-evidence-map-investigation-center.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-066: Evidence Map as Investigation Workspace Center'
 ---
 
diff --git a/docs/07-decisions/adr-067-unified-glm-regression.md b/docs/07-decisions/adr-067-unified-glm-regression.md
index cae511c4d..9fc9f5e5f 100644
--- a/docs/07-decisions/adr-067-unified-glm-regression.md
+++ b/docs/07-decisions/adr-067-unified-glm-regression.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-067: Unified GLM Regression Engine'
-status: accepted
+status: active
 date: 2026-04-05
 ---
 
diff --git a/docs/07-decisions/adr-068-coscout-cognitive-redesign.md b/docs/07-decisions/adr-068-coscout-cognitive-redesign.md
index 97f4c149e..eb6edfd09 100644
--- a/docs/07-decisions/adr-068-coscout-cognitive-redesign.md
+++ b/docs/07-decisions/adr-068-coscout-cognitive-redesign.md
@@ -1,13 +1,17 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-068: CoScout Cognitive Redesign'
-audience: [engineer]
+audience: human
 category: architecture
-status: stable
-related: [coscout, ai-integration, prompt-caching, tiered-prompts]
+status: active
+related: [coscout, ai-integration, prompt-caching, tiered-prompts, coscout-ax-design]
 ---
 
 # ADR-068: CoScout Cognitive Redesign
 
+> See canonical AX design at [coscout-ax-design](../01-vision/coscout-ax-design.md) for the consolidated persona, tier-gating, safety, and prompt engineering overview. This ADR is the decision provenance for the modular prompt architecture, tier model, and phase-adaptive assembly.
+
 **Status:** Accepted  
 **Date:** 2026-04-05  
 **Supersedes:** Accumulated prompt additions from ADR-019, 029, 047, 049, 053, 054, 060, 064, 065, 067
diff --git a/docs/07-decisions/adr-069-three-boundary-numeric-safety.md b/docs/07-decisions/adr-069-three-boundary-numeric-safety.md
index 0d4122ae9..741a8d88d 100644
--- a/docs/07-decisions/adr-069-three-boundary-numeric-safety.md
+++ b/docs/07-decisions/adr-069-three-boundary-numeric-safety.md
@@ -1,13 +1,17 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-069: Three-Boundary Numeric Safety'
-audience: [engineer]
+audience: human
 category: architecture
-status: stable
-related: [stats-engine, numerical-safety, nan-defense, safe-math]
+status: active
+related: [stats-engine, numerical-safety, nan-defense, safe-math, coscout-ax-design]
 ---
 
 # ADR-069: Three-Boundary Numeric Safety
 
+> See canonical AX design at [coscout-ax-design](../01-vision/coscout-ax-design.md) for the consolidated CoScout safety overview including REF marker constraints. This ADR is the decision provenance for the B1/B2/B3 numeric safety architecture and `safeMath.ts`.
+
 **Status:** Accepted
 **Date:** 2026-04-06
 
diff --git a/docs/07-decisions/adr-070-frame-workspace.md b/docs/07-decisions/adr-070-frame-workspace.md
index 9fd6449a1..819a77f7c 100644
--- a/docs/07-decisions/adr-070-frame-workspace.md
+++ b/docs/07-decisions/adr-070-frame-workspace.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-070: FRAME Workspace, Visual Process Map & Deterministic Mode Inference'
-audience: [engineer]
+audience: human
 category: architecture
-status: stable
+status: active
 related:
   [
     frame-workspace,
diff --git a/docs/07-decisions/adr-071-coscout-voice-input.md b/docs/07-decisions/adr-071-coscout-voice-input.md
index a63a72786..de1169d4f 100644
--- a/docs/07-decisions/adr-071-coscout-voice-input.md
+++ b/docs/07-decisions/adr-071-coscout-voice-input.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-071: CoScout Voice Input'
 ---
 
diff --git a/docs/07-decisions/adr-072-process-hub-storage-and-coscout-context.md b/docs/07-decisions/adr-072-process-hub-storage-and-coscout-context.md
index cfbff033f..20a2719a3 100644
--- a/docs/07-decisions/adr-072-process-hub-storage-and-coscout-context.md
+++ b/docs/07-decisions/adr-072-process-hub-storage-and-coscout-context.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-072: Process Hub Storage and CoScout Context Boundary'
-status: accepted
+status: active
 date: 2026-04-26
 ---
 
diff --git a/docs/07-decisions/adr-073-no-statistical-rollup-across-heterogeneous-units.md b/docs/07-decisions/adr-073-no-statistical-rollup-across-heterogeneous-units.md
index 649dce683..d0b44f310 100644
--- a/docs/07-decisions/adr-073-no-statistical-rollup-across-heterogeneous-units.md
+++ b/docs/07-decisions/adr-073-no-statistical-rollup-across-heterogeneous-units.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-073: No Statistical Roll-Up Across Heterogeneous Units'
-status: accepted
+status: active
 date: 2026-04-29
 ---
 
diff --git a/docs/07-decisions/adr-074-scout-level-spanning-surface-boundary-policy.md b/docs/07-decisions/adr-074-scout-level-spanning-surface-boundary-policy.md
index 5131a6868..e5eadd4ba 100644
--- a/docs/07-decisions/adr-074-scout-level-spanning-surface-boundary-policy.md
+++ b/docs/07-decisions/adr-074-scout-level-spanning-surface-boundary-policy.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-074: SCOUT level-spanning surface — boundary policy'
-audience: [product, designer, engineer, analyst]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-04-29
 related:
   - adr-073-no-statistical-rollup-across-heterogeneous-units
diff --git a/docs/07-decisions/adr-075-pwa-atomic-deploy-and-update-policy.md b/docs/07-decisions/adr-075-pwa-atomic-deploy-and-update-policy.md
index ab0217b9e..36a1df493 100644
--- a/docs/07-decisions/adr-075-pwa-atomic-deploy-and-update-policy.md
+++ b/docs/07-decisions/adr-075-pwa-atomic-deploy-and-update-policy.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-075: PWA atomic deploy + update policy'
-audience: [developer, architect]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-05-02
 related:
   - adr-058-deployment-lifecycle
diff --git a/docs/07-decisions/adr-076-frame-b0-lightweight-render.md b/docs/07-decisions/adr-076-frame-b0-lightweight-render.md
index f4d7f5322..59cf0ec08 100644
--- a/docs/07-decisions/adr-076-frame-b0-lightweight-render.md
+++ b/docs/07-decisions/adr-076-frame-b0-lightweight-render.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-076: FRAME b0 lightweight render — investigator vs author archetypes'
-audience: [product, designer, engineer, analyst]
+audience: human
 category: architecture
 status: superseded
 superseded-by: docs/superpowers/specs/2026-05-04-canvas-migration-design.md
diff --git a/docs/07-decisions/adr-077-snapshot-provenance-and-match-summary-wedge.md b/docs/07-decisions/adr-077-snapshot-provenance-and-match-summary-wedge.md
index d2c7d7b3e..a2b02b8a3 100644
--- a/docs/07-decisions/adr-077-snapshot-provenance-and-match-summary-wedge.md
+++ b/docs/07-decisions/adr-077-snapshot-provenance-and-match-summary-wedge.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-077: Snapshot-level provenance + match-summary wedge for paste-into-existing-Hub'
-audience: [product, designer, engineer, analyst]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-05-04
 related:
   - adr-073-no-statistical-rollup-across-heterogeneous-units
diff --git a/docs/07-decisions/adr-078-pwa-azure-architecture-alignment.md b/docs/07-decisions/adr-078-pwa-azure-architecture-alignment.md
index 5ac17da35..c2afac176 100644
--- a/docs/07-decisions/adr-078-pwa-azure-architecture-alignment.md
+++ b/docs/07-decisions/adr-078-pwa-azure-architecture-alignment.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-078: PWA + Azure architecture alignment — same product, gated tiers'
-audience: [product, designer, engineer]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-05-05
 related:
   - adr-012-pwa-browser-only
diff --git a/docs/07-decisions/adr-079-etag-optimistic-concurrency-paid-tier-hub-writes.md b/docs/07-decisions/adr-079-etag-optimistic-concurrency-paid-tier-hub-writes.md
index 9a9704114..1925a86ce 100644
--- a/docs/07-decisions/adr-079-etag-optimistic-concurrency-paid-tier-hub-writes.md
+++ b/docs/07-decisions/adr-079-etag-optimistic-concurrency-paid-tier-hub-writes.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-079: ETag optimistic concurrency for paid-tier hub-blob writes'
-audience: [product, designer, engineer]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-05-07
 related:
   - adr-058-deployment-lifecycle
diff --git a/docs/07-decisions/adr-080-sustainment-auto-fire-pattern.md b/docs/07-decisions/adr-080-sustainment-auto-fire-pattern.md
index a7d43ff60..bb935743e 100644
--- a/docs/07-decisions/adr-080-sustainment-auto-fire-pattern.md
+++ b/docs/07-decisions/adr-080-sustainment-auto-fire-pattern.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-080: Sustainment auto-fire + Inbox prompt + signoff-gated lifecycle pattern'
-audience: [product, designer, engineer]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-05-13
 related:
   - adr-064-suspected-cause-hub-model
diff --git a/docs/07-decisions/adr-081-canvas-viewport-architecture.md b/docs/07-decisions/adr-081-canvas-viewport-architecture.md
index 028f3be01..181ea2b94 100644
--- a/docs/07-decisions/adr-081-canvas-viewport-architecture.md
+++ b/docs/07-decisions/adr-081-canvas-viewport-architecture.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-081: Canvas viewport architecture — levels-as-pan/zoom (8f)'
-audience: [product, designer, engineer, analyst]
+audience: human
 category: architecture
-status: accepted
+status: active
 date: 2026-05-13
 related:
   - adr-068-coscout-cognitive-redesign
diff --git a/docs/07-decisions/adr-082-wedge-architecture.md b/docs/07-decisions/adr-082-wedge-architecture.md
index 5d85db816..34f863704 100644
--- a/docs/07-decisions/adr-082-wedge-architecture.md
+++ b/docs/07-decisions/adr-082-wedge-architecture.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'ADR-082: Wedge architecture — single-product VariScout'
 ---
 
diff --git a/docs/07-decisions/adr-083-eight-purpose-doc-taxonomy.md b/docs/07-decisions/adr-083-eight-purpose-doc-taxonomy.md
new file mode 100644
index 000000000..07001910c
--- /dev/null
+++ b/docs/07-decisions/adr-083-eight-purpose-doc-taxonomy.md
@@ -0,0 +1,171 @@
+---
+title: 'ADR-083: Eight-purpose doc taxonomy + four-tier velocity'
+status: active
+date: 2026-05-16
+purpose: decide
+tier: living
+audience: both
+topic: [docs-strategy, schema, meta]
+related: [2026-05-16-docs-strategy-design, 2026-05-16-docs-strategy-memo]
+---
+
+# ADR-083: Eight-purpose doc taxonomy + four-tier velocity
+
+**Status:** Accepted
+**Date:** 2026-05-16
+**Related:** [Docs Strategy Design](../superpowers/specs/2026-05-16-docs-strategy-design.md), [Docs Strategy Memo](../superpowers/specs/2026-05-16-docs-strategy-memo.md)
+
+## Context
+
+VariScout's frontmatter schema has accumulated 22 STATUS values, 18 CATEGORY values, and 14 AUDIENCE values across ~521 markdown files. The schema was built incrementally and now exhibits two failure modes simultaneously:
+
+1. **Over-discipline at the metadata layer** — 22 STATUS enums create aspirational distinctions (`review`, `living`, `raw`, `template`) that are never consistently applied and require constant maintenance. Most docs cycle between `draft → active → superseded` in practice.
+
+2. **Under-discipline at the lifecycle layer** — no freshness signal (`last-verified`), no implementation-drift sensor (`verified-against-commit`), no velocity-tier to drive maintenance cadence, no purpose field to route retrieval by function.
+
+The result: expensive enum schemas that don't prevent drift, while the fields that would detect drift are absent.
+
+An audit of 487 active docs (521 files total, 34 supporting README/index hubs) revealed:
+
+- Design + Decide dominate living-tier docs (51% + 25% = 76%)
+- **Agent-context is critically under-engineered at 4% of living docs** — the AX-dev surface that reads on every subagent dispatch
+- 17 emergent topic tags emerge naturally; formal CATEGORY enum enforces a 2024 ontology instead
+
+Additionally, industry documentation frameworks (Diátaxis, Google's engineering-practices) are human-reader-centric and pre-LLM-era. VariScout needs a taxonomy covering both human readers AND the AX-dev surface (subagents) that now constitutes half its documentation system.
+
+## Decision
+
+Replace the existing three-enum metadata schema with a two-axis schema: **8 purposes × 4 tiers**, with free-form topic tags and collapsed enums.
+
+### The new schema
+
+```yaml
+# Core two-axis classification
+purpose: orient | decide | design | build | system | constrain | agent-context | remember
+tier: stable | living | ephemeral | card
+
+# Collapsed enums
+status: draft | active | superseded | archived # was 22 values
+audience: human | agent | both # was 14 values
+topic: [<kebab-tags>] # replaces CATEGORY enum; free-form
+
+# New operational fields
+last-verified: YYYY-MM-DD # drift sensor
+verified-against-commit: <sha> # drift sensor
+supersedes: [<id>] # supersession chain
+related: [<id>] # cross-reference graph
+```
+
+### The 8 purposes
+
+| Purpose           | Reader intent                              | Failure mode                            |
+| ----------------- | ------------------------------------------ | --------------------------------------- |
+| **orient**        | "What is VariScout? Why? For whom?"        | Stale narrative misleads                |
+| **decide**        | "What did we conclude?"                    | Re-litigation; lost provenance          |
+| **design**        | "What is/should the product be?"           | Spec ↔ implementation drift             |
+| **build**         | "How do I implement this _here_?"          | Agent uses outdated procedure           |
+| **system**        | "How is the system architected?"           | Diagrams describe old architecture      |
+| **constrain**     | "What must I never do?"                    | Constraint ignored; not enforced        |
+| **agent-context** | "I'm an agent — what do I need right now?" | Context bloat; wrong information loaded |
+| **remember**      | "How did we get here? Why was X decided?"  | Lost institutional knowledge            |
+
+### The 4 tiers
+
+| Tier          | Volatility                              | Maintenance                      |
+| ------------- | --------------------------------------- | -------------------------------- |
+| **stable**    | Years-stable                            | Quarterly human review           |
+| **living**    | Months-stable, amend-not-rewrite        | Steward-loop drift checks weekly |
+| **ephemeral** | Days-to-weeks, lifecycle = until-merged | Auto-archive on close            |
+| **card**      | Atomic, append-only                     | Continuous (Steward + telemetry) |
+
+### STATUS alias map (transitional)
+
+Old values that no longer exist in the canonical enum, mapped to their replacements:
+
+| Old value     | → New value  | Rationale                                          |
+| ------------- | ------------ | -------------------------------------------------- |
+| `accepted`    | `active`     | ADR language; active is the general term           |
+| `in-progress` | `active`     | Work-in-progress is still active                   |
+| `delivered`   | `active`     | Delivered spec is still the active reference       |
+| `reference`   | `active`     | Reference docs are active by definition            |
+| `template`    | `active`     | Templates are active infrastructure                |
+| `review`      | `draft`      | Under review = still a draft                       |
+| `design`      | `draft`      | Design phase = draft                               |
+| `raw`         | `draft`      | Raw = draft                                        |
+| `stable`      | `active`     | Stable was a STATUS; now tier=stable captures this |
+| `living`      | `active`     | Living was a STATUS; now tier=living captures this |
+| `deferred`    | `superseded` | Deferred work is effectively superseded            |
+
+### AUDIENCE alias map (transitional)
+
+| Old value         | → New value |
+| ----------------- | ----------- |
+| `developer`       | `human`     |
+| `engineer`        | `human`     |
+| `analyst`         | `human`     |
+| `business`        | `human`     |
+| `architect`       | `human`     |
+| `admin`           | `human`     |
+| `designer`        | `human`     |
+| `product`         | `human`     |
+| `manager`         | `human`     |
+| `compliance`      | `human`     |
+| `procurement`     | `human`     |
+| `infosec`         | `human`     |
+| `quality-manager` | `human`     |
+| `auditor`         | `human`     |
+
+All current AUDIENCE values map to `human`; `agent` and `both` are new explicit values for the AX-dev surface.
+
+### CATEGORY replacement
+
+CATEGORY enum is removed entirely. Free-form `topic` tags replace it. Dominant tags from the audit:
+
+```
+adr, stats, charts, ux, projects, investigation, canvas, ax, coscout,
+methodology, azure, capability, response-paths, stores, marketplace, i18n, tier-gating
+```
+
+Any kebab-case value is valid. The enumeration emerges from usage rather than being prescribed upfront.
+
+## Migration
+
+**Play 1 transition cycle** (this PR):
+
+1. `scripts/docs-frontmatter-schema.mjs` — add `PURPOSE` + `TIER` enums, collapse STATUS to 4 + add alias map, add AUDIENCE alias map, remove CATEGORY enum, add new optional fields.
+2. `scripts/check-doc-frontmatter.mjs` — warn (not fail) when old STATUS/CATEGORY/AUDIENCE values appear via alias map; hard-fail on unknown PURPOSE/TIER when present; report missing `purpose`+`tier` counts.
+3. `scripts/docs-frontmatter-fix.mjs` — backfill `purpose`+`tier` via path heuristics; map old STATUS/AUDIENCE via alias maps.
+4. All docs `git mv`'d to new tier-based folder hierarchy.
+
+**Back-compatibility**: Old STATUS/AUDIENCE values warn but pass during the transition cycle (one release). After Play 2 completes, warnings become failures.
+
+**Folder restructure**: `docs/` moves from 11 numbered folders to `stable/`, `living/<purpose>/`, `ephemeral/`, `cards/`, `archive/`. Git history preserved via `git mv`.
+
+## Consequences
+
+**Positive**:
+
+- Adding a doc is cheaper: 2 required fields (`title`, `purpose`) instead of 4 (`title`, `audience`, `category`, `status`).
+- Drift is now detectable: `last-verified` + `verified-against-commit` enable the Steward loop (Play 2e).
+- Topic tags evolve with the product; enum doesn't lock 2024 ontology.
+- Agent-context gap becomes measurable (cohort report, Play 5).
+- AX-dev surface explicitly represented via `purpose: agent-context`.
+
+**Negative**:
+
+- Short-term migration friction on 521 files (mitigated by `docs-frontmatter-fix.mjs` automation).
+- Free-form topic tags require human discipline; no enum enforcement (mitigated by 17-tag vocabulary guidance).
+- `classify()` function in schema needs updating for new folder paths.
+
+**Neutral**:
+
+- ADR shape is preserved; ADRs just gain optional `purpose: decide` + `tier: living` fields.
+- Validation tooling (`scripts/check-doc-frontmatter.mjs`) is extended, not replaced.
+
+## Related artifacts
+
+- [2026-05-16-docs-strategy-design.md](../superpowers/specs/2026-05-16-docs-strategy-design.md) — full strategy spec
+- [2026-05-16-docs-strategy-memo.md](../superpowers/specs/2026-05-16-docs-strategy-memo.md) — CTO memo
+- `scripts/docs-frontmatter-schema.mjs` — schema SSOT (updated in Play 1)
+- `scripts/check-doc-frontmatter.mjs` — validation (updated in Play 1)
+- `scripts/docs-frontmatter-fix.mjs` — backfill automation (updated in Play 1)
diff --git a/docs/07-decisions/audit-2026-02-state-of-product.md b/docs/07-decisions/audit-2026-02-state-of-product.md
index e91e78dc8..3b3bcc886 100644
--- a/docs/07-decisions/audit-2026-02-state-of-product.md
+++ b/docs/07-decisions/audit-2026-02-state-of-product.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'State-of-the-Product Audit — February 2026'
 ---
 
diff --git a/docs/07-decisions/index.md b/docs/07-decisions/index.md
index 1aaab08b3..ad7394668 100644
--- a/docs/07-decisions/index.md
+++ b/docs/07-decisions/index.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: decide
 title: 'Architecture Decision Records'
 ---
 
diff --git a/docs/08-products/azure/ai-safety-report.md b/docs/08-products/azure/ai-safety-report.md
index 685f89d8b..faa51b0d9 100644
--- a/docs/08-products/azure/ai-safety-report.md
+++ b/docs/08-products/azure/ai-safety-report.md
@@ -1,9 +1,11 @@
 ---
+tier: stable
+purpose: orient
 title: AI Safety Report — CoScout
 category: compliance
-status: template
+status: active
 date: 2026-03-17
-audience: [admin, architect]
+audience: human
 ---
 
 # AI Safety Report — CoScout
diff --git a/docs/08-products/azure/arm-template.md b/docs/08-products/azure/arm-template.md
index d2050e932..cd1c5ed46 100644
--- a/docs/08-products/azure/arm-template.md
+++ b/docs/08-products/azure/arm-template.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'ARM Template Documentation'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # ARM Template Documentation
diff --git a/docs/08-products/azure/authentication.md b/docs/08-products/azure/authentication.md
index a59423dc2..1ab6fe582 100644
--- a/docs/08-products/azure/authentication.md
+++ b/docs/08-products/azure/authentication.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Authentication (EasyAuth)'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Authentication (EasyAuth)
diff --git a/docs/08-products/azure/blob-storage-sync.md b/docs/08-products/azure/blob-storage-sync.md
index bcb68c58d..82178628a 100644
--- a/docs/08-products/azure/blob-storage-sync.md
+++ b/docs/08-products/azure/blob-storage-sync.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Blob Storage Sync'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Blob Storage Sync
diff --git a/docs/08-products/azure/certification-guide.md b/docs/08-products/azure/certification-guide.md
index 68c5f8ef1..7003bb031 100644
--- a/docs/08-products/azure/certification-guide.md
+++ b/docs/08-products/azure/certification-guide.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Azure Marketplace Certification Guide'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Azure Marketplace Certification Guide
diff --git a/docs/08-products/azure/how-it-works.md b/docs/08-products/azure/how-it-works.md
index b834c7138..732df3051 100644
--- a/docs/08-products/azure/how-it-works.md
+++ b/docs/08-products/azure/how-it-works.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'How It Works'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # How It Works
diff --git a/docs/08-products/azure/index.md b/docs/08-products/azure/index.md
index 2108aa2e2..ea47e66d7 100644
--- a/docs/08-products/azure/index.md
+++ b/docs/08-products/azure/index.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Azure App (Primary Product)'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Azure App (Primary Product)
diff --git a/docs/08-products/azure/marketplace-readiness-report.md b/docs/08-products/azure/marketplace-readiness-report.md
index 9489fcd8a..54912e9f6 100644
--- a/docs/08-products/azure/marketplace-readiness-report.md
+++ b/docs/08-products/azure/marketplace-readiness-report.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: Marketplace Readiness Report
 date: 2026-03-25
-status: in-progress
-audience: [admin, architect]
+status: active
+audience: human
 category: reference
 ---
 
diff --git a/docs/08-products/azure/marketplace.md b/docs/08-products/azure/marketplace.md
index 6515fd5e1..2bb918bec 100644
--- a/docs/08-products/azure/marketplace.md
+++ b/docs/08-products/azure/marketplace.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Azure Marketplace Guide'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Azure Marketplace Guide
diff --git a/docs/08-products/azure/pricing-tiers.md b/docs/08-products/azure/pricing-tiers.md
index fa174e1b3..a1e1a2242 100644
--- a/docs/08-products/azure/pricing-tiers.md
+++ b/docs/08-products/azure/pricing-tiers.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Pricing'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Pricing
diff --git a/docs/08-products/azure/security-whitepaper.md b/docs/08-products/azure/security-whitepaper.md
index 59ca5bb34..0d2a3eebd 100644
--- a/docs/08-products/azure/security-whitepaper.md
+++ b/docs/08-products/azure/security-whitepaper.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Security Whitepaper'
-audience: [procurement, infosec]
+audience: human
 category: compliance
-status: stable
+status: active
 date: 2026-04-03
 ---
 
diff --git a/docs/08-products/azure/storage.md b/docs/08-products/azure/storage.md
index ac9c2d579..8c5d44d73 100644
--- a/docs/08-products/azure/storage.md
+++ b/docs/08-products/azure/storage.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Azure App Storage'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Azure App Storage
diff --git a/docs/08-products/azure/submission-checklist.md b/docs/08-products/azure/submission-checklist.md
index e36d4dcd4..bc4356c65 100644
--- a/docs/08-products/azure/submission-checklist.md
+++ b/docs/08-products/azure/submission-checklist.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Azure Marketplace Submission Checklist'
-audience: [admin, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # Azure Marketplace Submission Checklist
diff --git a/docs/08-products/feature-parity.md b/docs/08-products/feature-parity.md
index 09060dc97..7d3be1fd8 100644
--- a/docs/08-products/feature-parity.md
+++ b/docs/08-products/feature-parity.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Feature Parity Matrix'
-audience: [business, product]
+audience: human
 category: strategy
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [tiers, wedge, pricing, modes]
 ---
diff --git a/docs/08-products/index.md b/docs/08-products/index.md
index 8ae390ec4..fe644b0fd 100644
--- a/docs/08-products/index.md
+++ b/docs/08-products/index.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Products'
-audience: [business, product]
+audience: human
 category: strategy
-status: stable
+status: active
 ---
 
 > **⚠️ Queued for wedge rewrite (2026-05-16)** — see [wedge spec](../superpowers/specs/2026-05-16-wedge-architecture-design.md) + [ADR-082](../07-decisions/adr-082-wedge-architecture.md) for V1 canonical anatomy. tier/product summary collapses to PWA + €99 single SKU; VariScout Process noted as future product. This doc has a wedge-marker header only; full content rewrite is queued for Phase C.3 batch 4 (subagent-driven).
diff --git a/docs/08-products/iso-9001-alignment.md b/docs/08-products/iso-9001-alignment.md
index fa7740b9b..a3c7e4757 100644
--- a/docs/08-products/iso-9001-alignment.md
+++ b/docs/08-products/iso-9001-alignment.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'ISO 9001:2026 Alignment Guide'
-audience: [quality-manager, auditor]
+audience: human
 category: compliance
-status: stable
+status: active
 date: 2026-04-03
 ---
 
diff --git a/docs/08-products/presentations/variscout-product-evaluation-narrative-plan.md b/docs/08-products/presentations/variscout-product-evaluation-narrative-plan.md
index ef1a4c926..cbd2952c3 100644
--- a/docs/08-products/presentations/variscout-product-evaluation-narrative-plan.md
+++ b/docs/08-products/presentations/variscout-product-evaluation-narrative-plan.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: orient
 title: VariScout Product Evaluation Deck Narrative Plan
-audience: [product, manager, analyst]
+audience: human
 category: strategy
 status: draft
 date: 2026-04-27
diff --git a/docs/08-products/pwa/index.md b/docs/08-products/pwa/index.md
index 7f1311d00..ec4dbcd22 100644
--- a/docs/08-products/pwa/index.md
+++ b/docs/08-products/pwa/index.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'PWA (Free Training Tool)'
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # PWA (Free Training Tool)
diff --git a/docs/08-products/tier-philosophy.md b/docs/08-products/tier-philosophy.md
index 1d5dc46a8..dadc8d489 100644
--- a/docs/08-products/tier-philosophy.md
+++ b/docs/08-products/tier-philosophy.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: Tier Philosophy
-audience: [analyst, engineer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [pricing, tiers, feature-gating, value-ladder]
 ---
 
diff --git a/docs/08-products/website/content-architecture.md b/docs/08-products/website/content-architecture.md
index f129675d0..81e630ae1 100644
--- a/docs/08-products/website/content-architecture.md
+++ b/docs/08-products/website/content-architecture.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Website Content Architecture'
-audience: [business, designer]
+audience: human
 category: strategy
-status: stable
+status: active
 ---
 
 # Website Content Architecture
diff --git a/docs/08-products/website/design-philosophy.md b/docs/08-products/website/design-philosophy.md
index b9a3d49f0..459de7c48 100644
--- a/docs/08-products/website/design-philosophy.md
+++ b/docs/08-products/website/design-philosophy.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Website Design Philosophy'
-audience: [business, designer]
+audience: human
 category: strategy
-status: stable
+status: active
 ---
 
 # Website Design Philosophy
diff --git a/docs/08-products/website/index.md b/docs/08-products/website/index.md
index 40426a4f2..009378cd1 100644
--- a/docs/08-products/website/index.md
+++ b/docs/08-products/website/index.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Marketing Website'
-audience: [business, designer]
+audience: human
 category: strategy
-status: stable
+status: active
 ---
 
 # Marketing Website
diff --git a/docs/09-baseline/2026-04-17-agent-docs-baseline.md b/docs/09-baseline/2026-04-17-agent-docs-baseline.md
index ac7c2ba9b..08df46592 100644
--- a/docs/09-baseline/2026-04-17-agent-docs-baseline.md
+++ b/docs/09-baseline/2026-04-17-agent-docs-baseline.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Agent Docs Architecture Baseline (Pre-Phase 1)
-audience: [engineer]
+audience: human
 category: reference
-status: reference
+status: active
 last-reviewed: 2026-04-17
 related: [claude-md, skills, migration, measurement]
 ---
diff --git a/docs/09-baseline/2026-04-17-doc-audit.md b/docs/09-baseline/2026-04-17-doc-audit.md
index b48da59ae..077e5f461 100644
--- a/docs/09-baseline/2026-04-17-doc-audit.md
+++ b/docs/09-baseline/2026-04-17-doc-audit.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: 'Doc Prune — Load-Bearing Audit Output'
-audience: [engineer, architect]
+audience: human
 category: architecture
 status: draft
 related: [doc-prune-audit, agent-docs-architecture]
diff --git a/docs/09-tutorials/index.md b/docs/09-tutorials/index.md
index 1ab8a57b8..effa5610e 100644
--- a/docs/09-tutorials/index.md
+++ b/docs/09-tutorials/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Tutorials
-audience: [analyst, engineer]
+audience: human
 category: tutorial
-status: deferred
+status: superseded
 related: [getting-started, learning, case-studies]
 ---
 
diff --git a/docs/10-development/discussions/2026-03-29-factor-intelligence-design.md b/docs/10-development/discussions/2026-03-29-factor-intelligence-design.md
index 36b5361b2..9eb76fcf1 100644
--- a/docs/10-development/discussions/2026-03-29-factor-intelligence-design.md
+++ b/docs/10-development/discussions/2026-03-29-factor-intelligence-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: build
 title: 'Factor Intelligence Design Discussion'
-audience: [developer, product]
+audience: human
 category: reference
 status: draft
 ---
diff --git a/docs/10-development/discussions/2026-03-29-probability-plot-analysis.md b/docs/10-development/discussions/2026-03-29-probability-plot-analysis.md
index 9b4a0286a..63f5726bc 100644
--- a/docs/10-development/discussions/2026-03-29-probability-plot-analysis.md
+++ b/docs/10-development/discussions/2026-03-29-probability-plot-analysis.md
@@ -1,10 +1,12 @@
 ---
+tier: living
+purpose: build
 title: Probability Plot — Multi-Angle Analysis
 date: 2026-03-29
 sources:
   - 2026-03-29-probability-plot.txt
   - 2026-03-29-probability-plot-and-commenting.txt
-audience: [developer, product]
+audience: human
 category: reference
 status: draft
 ---
diff --git a/docs/10-development/feature-backlog.md b/docs/10-development/feature-backlog.md
index 294925ed4..703fda7d7 100644
--- a/docs/10-development/feature-backlog.md
+++ b/docs/10-development/feature-backlog.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: build
 title: Feature Backlog
-audience: [developer, product]
+audience: human
 category: reference
 status: draft
 ---
diff --git a/docs/10-development/index.md b/docs/10-development/index.md
index 02c25c845..43117efdf 100644
--- a/docs/10-development/index.md
+++ b/docs/10-development/index.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: build
 title: Development
-audience: [developer, product]
+audience: human
 category: reference
 status: draft
 ---
diff --git a/docs/10-development/project-status-audit-2026-04-16.md b/docs/10-development/project-status-audit-2026-04-16.md
index f9e2e19ce..a417bad29 100644
--- a/docs/10-development/project-status-audit-2026-04-16.md
+++ b/docs/10-development/project-status-audit-2026-04-16.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: build
 title: 'Project Status Audit — 2026-04-16'
-audience: [engineer, developer]
+audience: human
 category: reference
-status: stable
+status: active
 related: [audit, state-of-product, adr, specs, feature-parity]
 ---
 
diff --git a/docs/10-development/transcripts/best-subset-regression-approach.md b/docs/10-development/transcripts/best-subset-regression-approach.md
index 429283786..c34d6af90 100644
--- a/docs/10-development/transcripts/best-subset-regression-approach.md
+++ b/docs/10-development/transcripts/best-subset-regression-approach.md
@@ -1,11 +1,13 @@
 ---
+tier: living
+purpose: build
 title: Best Subset Regression Approach
 type: transcript
 language: fi
 date: 2026-04-05
 duration: 199.7s
-status: raw
-audience: [developer, product]
+status: draft
+audience: human
 category: reference
 ---
 
diff --git a/docs/10-development/transcripts/dataset-analysis-discussion.md b/docs/10-development/transcripts/dataset-analysis-discussion.md
index 1a7350a43..6fa036882 100644
--- a/docs/10-development/transcripts/dataset-analysis-discussion.md
+++ b/docs/10-development/transcripts/dataset-analysis-discussion.md
@@ -1,11 +1,13 @@
 ---
+tier: living
+purpose: build
 title: Dataset Analysis Discussion
 type: transcript
 language: en
 date: 2026-03-29
 duration: 239.9s
-status: raw
-audience: [developer, product]
+status: draft
+audience: human
 category: reference
 ---
 
diff --git a/docs/10-development/transcripts/index.md b/docs/10-development/transcripts/index.md
index cf5346ae7..a1396f072 100644
--- a/docs/10-development/transcripts/index.md
+++ b/docs/10-development/transcripts/index.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: build
 title: Transcripts & Expert Interviews
-audience: [developer, product]
+audience: human
 category: reference
-status: raw
+status: draft
 ---
 
 # Transcripts & Expert Interviews
diff --git a/docs/10-development/transcripts/mbb-investigation-spine-validation.md b/docs/10-development/transcripts/mbb-investigation-spine-validation.md
index 7eaa01cb8..9ca9657a6 100644
--- a/docs/10-development/transcripts/mbb-investigation-spine-validation.md
+++ b/docs/10-development/transcripts/mbb-investigation-spine-validation.md
@@ -1,10 +1,12 @@
 ---
+tier: living
+purpose: build
 title: MBB Investigation Spine Validation
 type: interview
 language: en
 date: 2026-03-29
-status: delivered
-audience: [developer, product]
+status: active
+audience: human
 category: reference
 ---
 
diff --git a/docs/10-development/transcripts/probability-plot-in-analysis-flow.md b/docs/10-development/transcripts/probability-plot-in-analysis-flow.md
index bddc1aec3..e63b389d4 100644
--- a/docs/10-development/transcripts/probability-plot-in-analysis-flow.md
+++ b/docs/10-development/transcripts/probability-plot-in-analysis-flow.md
@@ -1,11 +1,13 @@
 ---
+tier: living
+purpose: build
 title: Probability Plot in Analysis Flow
 type: transcript
 language: en
 date: 2026-04-05
 duration: 359.2s
-status: raw
-audience: [developer, product]
+status: draft
+audience: human
 category: reference
 ---
 
diff --git a/docs/10-development/transcripts/projects-at-hp.md b/docs/10-development/transcripts/projects-at-hp.md
index 8ab5f94f0..ee9894cb6 100644
--- a/docs/10-development/transcripts/projects-at-hp.md
+++ b/docs/10-development/transcripts/projects-at-hp.md
@@ -1,11 +1,13 @@
 ---
+tier: living
+purpose: build
 title: Projects at HP
 type: transcript
 language: en
 date: 2026-04-05
 duration: 1905.8s
-status: raw
-audience: [developer, product]
+status: draft
+audience: human
 category: reference
 ---
 
diff --git a/docs/DATA-FLOW.md b/docs/DATA-FLOW.md
index b20c6eed6..b927774bc 100644
--- a/docs/DATA-FLOW.md
+++ b/docs/DATA-FLOW.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: system
 title: VariScout Data Lifecycle
-audience: [engineer]
+audience: human
 category: architecture
-status: stable
+status: active
 last-reviewed: 2026-04-24
 related: [data-flow, persistence, sync, customer-owned, blob-storage, indexeddb, easyauth]
 ---
diff --git a/docs/OVERVIEW.md b/docs/OVERVIEW.md
index 16ae15728..d5829de18 100644
--- a/docs/OVERVIEW.md
+++ b/docs/OVERVIEW.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: VariScout — What It Does In Practice
-audience: [engineer, analyst]
+audience: human
 category: reference
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [product-overview, modes, tiers, coscout, journey, wedge]
 ---
@@ -20,16 +22,7 @@ Canonical V1 design lives in the [wedge architecture spec](superpowers/specs/202
 
 ## The journey model
 
-Every investigation — whether quick or project-anchored — follows one methodological spine:
-
-**FRAME → SCOUT → INVESTIGATE → IMPROVE**
-
-- **FRAME.** State the problem (data-first or hypothesis-first entry). CoScout helps articulate.
-- **SCOUT.** Data is parsed and characterized. Four Lenses of variation emerge (central tendency, spread, pattern, distribution).
-- **INVESTIGATE.** Pick suspected causes — data-derived, gemba-observed, or expert-supplied — and examine each with the Evidence Map, statistics, and targeted questions. The Investigation Wall accumulates Findings linked to Hypotheses; Measurement Plans capture what evidence still needs collection (hypothesis-first path).
-- **IMPROVE.** Hypotheses converge on improvement actions. Inside a Project this becomes the Improve stage (action tracker), then Sustainment ("did it work? + close").
-
-The spine never changes. Analysis modes vary the tools used inside each phase.
+Every investigation — quick or project-anchored — follows one methodological spine: **FRAME → SCOUT → INVESTIGATE → IMPROVE** (state the problem; characterize the data with the Four Lenses; examine suspected causes via the Investigation Wall; converge on actions and verify in Sustainment). The spine never changes; analysis modes vary the tools inside each phase — see [USER-JOURNEYS.md](USER-JOURNEYS.md) for the detailed walk-through per mode and [methodology.md](01-vision/methodology.md) for the pedagogical narrative.
 
 ## Project — the optional formal wrapper
 
diff --git a/docs/USER-JOURNEYS-CAPABILITY.md b/docs/USER-JOURNEYS-CAPABILITY.md
index b1d101d16..2f8e29795 100644
--- a/docs/USER-JOURNEYS-CAPABILITY.md
+++ b/docs/USER-JOURNEYS-CAPABILITY.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Capability Mode — User Journey
-audience: [engineer, analyst]
+audience: human
 category: reference
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [capability, journey, cpk, spc, subgroup, wedge]
 ---
diff --git a/docs/USER-JOURNEYS-DEFECT.md b/docs/USER-JOURNEYS-DEFECT.md
index 642d4d715..7271d30ea 100644
--- a/docs/USER-JOURNEYS-DEFECT.md
+++ b/docs/USER-JOURNEYS-DEFECT.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Defect Analysis Mode — User Journey
-audience: [engineer, analyst]
+audience: human
 category: reference
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [defect, journey, ppm, event-rates, copq, wedge]
 ---
diff --git a/docs/USER-JOURNEYS-PERFORMANCE.md b/docs/USER-JOURNEYS-PERFORMANCE.md
index c3dc59ff7..cb3fc85e5 100644
--- a/docs/USER-JOURNEYS-PERFORMANCE.md
+++ b/docs/USER-JOURNEYS-PERFORMANCE.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Performance Mode — User Journey
-audience: [engineer, analyst]
+audience: human
 category: reference
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [performance, journey, multi-channel, cpk, drill-down, wedge]
 ---
diff --git a/docs/USER-JOURNEYS-PROCESS-FLOW.md b/docs/USER-JOURNEYS-PROCESS-FLOW.md
index 13ea98cea..8d612370d 100644
--- a/docs/USER-JOURNEYS-PROCESS-FLOW.md
+++ b/docs/USER-JOURNEYS-PROCESS-FLOW.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Process Flow Mode — User Journey
-audience: [engineer, analyst]
+audience: human
 category: reference
-status: design
+status: draft
 last-reviewed: 2026-05-16
 related: [process-flow, journey, bottleneck, design-only, wedge]
 ---
diff --git a/docs/USER-JOURNEYS-YAMAZUMI.md b/docs/USER-JOURNEYS-YAMAZUMI.md
index 38e132e39..3cc7b05a2 100644
--- a/docs/USER-JOURNEYS-YAMAZUMI.md
+++ b/docs/USER-JOURNEYS-YAMAZUMI.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Yamazumi Mode — User Journey
-audience: [engineer, analyst]
+audience: human
 category: reference
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [yamazumi, journey, lean, cycle-time, takt, wedge]
 ---
diff --git a/docs/USER-JOURNEYS.md b/docs/USER-JOURNEYS.md
index 371ff054b..8e67fa830 100644
--- a/docs/USER-JOURNEYS.md
+++ b/docs/USER-JOURNEYS.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: VariScout User Journeys — V1 Single-Persona Spine
-audience: [engineer, analyst]
+audience: human
 category: reference
-status: stable
+status: active
 last-reviewed: 2026-05-16
 related: [personas, flows, journey, modes, wedge]
 ---
diff --git a/docs/agent-context/doc-discipline.md b/docs/agent-context/doc-discipline.md
new file mode 100644
index 000000000..ad2b6b923
--- /dev/null
+++ b/docs/agent-context/doc-discipline.md
@@ -0,0 +1,265 @@
+---
+title: 'Doc Discipline — SSoT by Doc Type'
+purpose: agent-context
+tier: living
+audience: agent
+topic: [ax, discipline, governance]
+status: active
+last-verified: 2026-05-17
+related: [adr-083-eight-purpose-doc-taxonomy, 2026-05-16-docs-strategy-design]
+---
+
+# Doc Discipline — Single Source of Truth by Doc Type
+
+> Loaded by the `agent-context-quickstart` skill. **Read before editing ANY canonical doc.**
+> Full design rationale: [`docs-strategy-design.md §2.7`](../superpowers/specs/2026-05-16-docs-strategy-design.md).
+
+## Core principles (in priority order)
+
+1. **Doc-type discipline.** Different doc types serve different jobs and update differently (see table below). Conflating creates orphan-amendment drift.
+2. **One canonical home per concept.** Every concept has exactly ONE authoritative doc that owns it. Other docs link, never restate. When the canonical changes, all linkers point to the same updated truth.
+3. **Reader-first banners.** Frontmatter is machine-readable; banners are human/agent-readable. When state diverges from "body is current truth" (superseded, materially edited, delivered, ADR with amendments below), say so at the top — first thing the reader sees.
+4. **Decision-log as temporal index.** Every canonical edit produces a decision-log entry in a standard format (see below). Canonical docs are "as of now"; the log is "what changed when and why".
+
+The wedge-amendment incident (2026-05-17) was a violation of principles 1, 2, and 3 simultaneously — a side-amendment spec was created (violated 1: design specs edit in place) that forked canonical state (violated 2: one canonical home) with no banner on the original (violated 3: no reader signal). Convention alone doesn't prevent this; the validator in Play 2b enforces mechanically.
+
+## The mechanism
+
+Each principle maps to mechanical enforcement (Play 2b):
+
+| Principle                   | Enforcement                                                                                                    |
+| --------------------------- | -------------------------------------------------------------------------------------------------------------- |
+| Doc-type discipline         | Validator HARD-FAILs `*-amendment-*.md` etc. filenames; WARNs on design-spec `## Amendment` heading additions  |
+| One canonical home          | Play 4 audit + consolidation; `supersedes:` chain validation                                                   |
+| Reader-first banners        | Validator HARD-FAILs `status: superseded` without banner; WARNs on `status: delivered` without `delivered-by:` |
+| Decision-log temporal index | Validator WARNs on git diff to decision-log lines older than 7 days                                            |
+
+## Doc types + update patterns
+
+| Doc type            | Where it lives                                                             | Job                                                  | Update pattern                                                                                                                          |
+| ------------------- | -------------------------------------------------------------------------- | ---------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------- |
+| **Design spec**     | `docs/superpowers/specs/`, `docs/01-vision/coscout-ax-design.md`, etc.     | Always-current intended state                        | **Edit in place.** Spec body = the truth at any moment.                                                                                 |
+| **ADR**             | `docs/07-decisions/adr-*.md`                                               | Point-in-time decision record                        | `## Amendment — YYYY-MM-DD` block at bottom. New ADR for fundamental reframings.                                                        |
+| **Decision-log**    | `docs/decision-log.md`                                                     | Chronological pinned-decisions index                 | Append-only. Never edit prior entries. New entry supersedes old.                                                                        |
+| **Generated docs**  | `*-generated.md`                                                           | Mechanical projection of source                      | Never hand-edit. Update the generator script + regenerate. `.prettierignore` excludes.                                                  |
+| **Plan files**      | `~/.claude/plans/`                                                         | Ephemeral session transcripts                        | Live private. Never become canonical. Optionally promote to `docs/ephemeral/transcripts/` for landmark sessions.                        |
+| **Investigations**  | `docs/investigations.md`                                                   | Pre-decision observations                            | Open entries editable while open. `[RESOLVED]` entries immutable.                                                                       |
+| **Memory**          | `~/.claude/projects/.../memory/`                                           | Cross-session durable facts for Claude               | Topic files: edit in place. Index in MEMORY.md: one-line entries ≤200 chars.                                                            |
+| **Agent manifests** | Root `CLAUDE.md`, `AGENTS.md`, `docs/llms.txt`, nested package `CLAUDE.md` | Agent-loaded "first read" context for every dispatch | Edit in place. CLAUDE.md size budget per file enforced (`scripts/check-claude-md-size.sh`). Keep tight — every agent reads these first. |
+
+## Anti-patterns (mechanically forbidden)
+
+| Anti-pattern                                                                                                               | Why wrong                                                                          | Right pattern                                                                                     |
+| -------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------- |
+| `*-amendment-design.md`, `*-revision-design.md`, `*-update-design.md`, `*-followup-design.md` in `docs/superpowers/specs/` | Orphan amendments invisible to a new session reading only the canonical            | Edit canonical in place + decision-log entry. **HARD-FAILed by validator (Play 2b).**             |
+| Editing ADR body after `status: accepted` (changing the decision itself)                                                   | Loses original decision context; future readers can't see what was decided         | New `## Amendment — YYYY-MM-DD` block at bottom, OR new ADR with `supersedes:` chain              |
+| Editing prior decision-log entries                                                                                         | Breaks chronological record                                                        | New entry with `[supersedes <prior-date>]` marker                                                 |
+| Hand-editing `*-generated.md` files                                                                                        | Out-of-sync with source; next regen wipes edits                                    | Update the generator script + regenerate                                                          |
+| Putting active-decision content only in plan files                                                                         | Plan files live in `~/.claude/plans/`, not in repo; future sessions can't see them | Land decisions in canonical home (ADR / spec / decision-log) before merging the PR they came from |
+| Renaming canonical docs without `supersedes:` chain                                                                        | Breaks inbound links + obscures lineage                                            | `git mv` + add `supersedes:` to new path's frontmatter + redirect banner at old                   |
+| Two design specs claiming canonical status for the same concept                                                            | Forks the SoT; readers don't know which to trust                                   | One canonical home per concept (Play 4 enforces). Other docs link, not restate.                   |
+
+## Spec lifecycle states (explicit)
+
+For design specs (`docs/superpowers/specs/` and equivalents), `status:` reflects reality:
+
+| Status       | Meaning                                          | Body update pattern                                                         |
+| ------------ | ------------------------------------------------ | --------------------------------------------------------------------------- |
+| `draft`      | Designing; not yet building                      | Edit freely. `last-verified` discipline starts at `active`.                 |
+| `active`     | Designed; build underway                         | Edit in place when intent changes. Bump `last-verified`.                    |
+| `delivered`  | Shipped; `delivered-by: PR #N` set               | Body describes current state of shipped feature. Edit when feature evolves. |
+| `superseded` | Replaced by another spec via `supersedes:` chain | Frozen. Redirect banner at top pointing to successor.                       |
+| `archived`   | Historical reference only                        | Frozen. Moved to `docs/archive/`. Not cited from canonical surfaces.        |
+
+Transitions are explicit + auditable (frontmatter diff is the signal).
+
+## Edit-in-place mechanics (for design specs)
+
+When editing a canonical spec mid-flight:
+
+1. **Make the edit** in place. Don't create a side file.
+2. **Update `last-verified: YYYY-MM-DD`** frontmatter to today.
+3. **Add decision-log entry** citing the change:
+   ```
+   - YYYY-MM-DD — <short title>. Spec edit: <doc>#<section> [supersedes prior text].
+     <Why>: <one-sentence rationale>. Commit: <sha>.
+   ```
+4. **Commit** with message covering both the spec edit and the log entry.
+5. **(If during build PR)** flag the change in the PR description so reviewers see it.
+
+## Reader-first banners at the top
+
+**Frontmatter alone is not enough.** Frontmatter (`status:`, `supersedes:`, `last-verified:`) is for tooling — machine-readable. Readers (humans + agents) need a **human-readable banner at the top** of the body, immediately after the H1.
+
+Industry convention: status at top, amendments at bottom (for ADRs only).
+
+### When a banner is required
+
+| Condition                                                                    | Banner required at top                                         |
+| ---------------------------------------------------------------------------- | -------------------------------------------------------------- |
+| `status: superseded`                                                         | YES — redirect to successor; tell reader "this is not current" |
+| `status: archived`                                                           | YES — historical reference only                                |
+| `supersedes: [<id>]` set + recent supersession                               | YES — note what this replaces, link the predecessor            |
+| Material in-place edit to a `status: active` design spec in the last 30 days | YES — note what section changed + link decision-log            |
+| `status: delivered` with `delivered-by: PR #N`                               | RECOMMENDED — note shipped state + link the PR                 |
+| ADR with one or more `## Amendment` blocks                                   | YES — note in the top status line that amendments exist below  |
+| Fresh `status: draft` or unchanged `status: active`                          | NOT required — body speaks for itself                          |
+
+### Banner templates
+
+**Superseded spec** (after H1):
+
+```
+> ⚠️ **SUPERSEDED 2026-MM-DD** by [`<new-spec>`](path). Preserved for historical reference.
+> For current intent, see the successor. Reason: <one-line rationale>.
+```
+
+**Spec with recent material in-place edit**:
+
+```
+> 🔄 **Last material edit 2026-MM-DD** — §<section> rewritten: <one-sentence what>.
+> See `decision-log` entry for the matching date for rationale.
+```
+
+**Active build**:
+
+```
+> 🟡 **Active build** — implementation underway via [`PR #N`](https://github.com/.../pull/N).
+> Spec body reflects current intent; verify against PR for delivery progress.
+```
+
+**Delivered**:
+
+```
+> ✅ **Delivered 2026-MM-DD** via [`PR #N`](url). Body describes shipped state; edit when feature evolves.
+> Code: `packages/<...>`, `apps/<...>`.
+```
+
+**ADR with amendments**:
+
+```
+> ✅ **Accepted 2026-MM-DD** | Amended 2026-MM-DD, 2026-MM-DD (see Amendments at bottom).
+```
+
+**Archived**:
+
+```
+> 🗄 **Archived 2026-MM-DD**. Historical reference only. Successor: [`<doc>`](path) or N/A.
+```
+
+### Why this matters (the wedge-amendment incident)
+
+When the wedge spec's §3.1 was amended on 2026-05-16, the canonical spec did NOT get a `🔄 Last material edit` banner. A fresh session reading the wedge spec saw no signal — body read as canonical truth, §3.1 said "Improve removed", subagent went down the wrong path. The amendment file existed (`2026-05-16-improve-tab-amendment-design.md`) but was invisible without active discovery.
+
+A banner at the top of the wedge spec would have made the divergence immediately visible to any reader.
+
+### Wikilinks don't substitute for banners
+
+`[[name]]` references are bidirectional in the graph but **passive at read time**. A reader sees them only if they:
+
+- Navigate the link (active)
+- Check a backlinks panel (requires tool support + active intent)
+- Run `pnpm docs:related <id>` (when Play 2 toolbox ships)
+
+Banners are **active** — the reader sees them whether they want to or not. Use banners for "you must know this before reading the body" signals; use wikilinks for "you may want to explore this".
+
+### Banner + frontmatter both
+
+The rule: if the frontmatter signals a condition that affects reader interpretation, the banner repeats it in human-readable form. Don't pick one over the other.
+
+| Information             | Frontmatter (tooling)                                       | Banner (reader)                                 |
+| ----------------------- | ----------------------------------------------------------- | ----------------------------------------------- |
+| Doc is superseded       | `status: superseded` + `supersedes: [...]`                  | "⚠️ SUPERSEDED" + link                          |
+| Doc was recently edited | `last-verified: <today>` + `verified-against-commit: <sha>` | "🔄 Last material edit" if user-relevant change |
+| Doc is delivered        | `status: delivered` + `delivered-by: PR #N`                 | "✅ Delivered" + PR link + code paths           |
+
+### Enforcement (Play 2b validator extensions)
+
+When Play 2b ships, the validator will (in addition to the filename anti-pattern check):
+
+- HARD-FAIL: `status: superseded` without `> SUPERSEDED` banner in first 10 lines of body
+- HARD-FAIL: `supersedes: [<id>]` without banner mentioning what's being superseded
+- WARN: `status: delivered` without `> Delivered` banner or `delivered-by:` frontmatter
+- WARN: design spec with `last-verified` >30 days behind current HEAD + status `active` (suggests freshness banner OR transition to `delivered`)
+
+---
+
+## Decision-log as temporal index
+
+Every change to a canonical doc — spec body edit, ADR amendment, supersession — gets a decision-log entry. The decision-log is the chronological backbone over the canonical doc set: canonical docs themselves are always "as of now"; the log tells you "what changed when and why".
+
+The `[supersedes <doc>#<section>]` marker is machine-readable + greppable for finding what changed.
+
+### Standard entry format
+
+Use this shape for every decision-log entry (Play 2c's `pnpm docs:recent --amendments` parses it):
+
+```
+- YYYY-MM-DD — <short title ≤80 chars>. <Edit type>: <doc>#<section> [supersedes <prior>].
+  Why: <one-sentence rationale>.
+  Commit: <sha-short>. PR: #N. Related: [[<id>], [<id>]].
+```
+
+**Required**: date, title, edit-type (`spec edit` / `ADR amendment` / `new ADR` / `supersession` / `archived` / `new spec`), `[supersedes …]` marker if applicable, Why, Commit.
+**Optional**: PR number, Related wikilinks to other ADRs/specs/entries.
+
+**Edit-type vocabulary** (use these strings exactly so the parser can categorize):
+
+| Edit type       | When to use                                                   |
+| --------------- | ------------------------------------------------------------- |
+| `spec edit`     | Material in-place change to a design spec body                |
+| `ADR amendment` | Appended `## Amendment` block to an existing ADR              |
+| `new ADR`       | Created a new ADR (with `supersedes:` if replacing)           |
+| `supersession`  | Marked an existing doc `status: superseded` + added successor |
+| `archived`      | Moved a superseded doc to `docs/archive/`                     |
+| `new spec`      | Created a new design spec (no supersedes)                     |
+
+**What NOT to put in entries**: full rationale paragraphs (those live in the brainstorm transcript or PR description that Why links to); restated spec content (canonical spec is the SoT); duplicated information across entries (one decision = one entry).
+
+## Decision tree: "I need to change a canonical doc"
+
+```
+Q1: Is the doc a design spec, ADR, or decision-log entry?
+
+  Design spec:
+    Q2: Is the change material reframing (the spec describes a fundamentally
+        different design now)?
+      Yes → New spec with supersedes: [<old-id>]. Mark old superseded.
+      No  → Edit the spec body in place. Add decision-log entry. Bump
+            last-verified. (DO NOT create *-amendment-*.md side file.)
+
+  ADR:
+    Q2: Is the change material reframing (the original decision is reversed
+        or fundamentally restructured)?
+      Yes → New ADR with supersedes: [<old-id>]. Mark old superseded.
+      No  → Append ## Amendment — YYYY-MM-DD block at the bottom of the ADR.
+            Add decision-log entry. (DO NOT edit the original decision text.)
+
+  Decision-log entry:
+    Always → Append a NEW entry with [supersedes <prior-date>] marker.
+             (DO NOT edit the prior entry.)
+```
+
+## Why this matters
+
+Without doc-type discipline, the 8-purpose × 4-tier schema is just metadata gymnastics. With it, the corpus stays coherent as it scales. At any moment, "what is the current intended state of X?" has exactly ONE answer: the canonical doc for X. No mental merging. No side-spec hunting. No "what supersedes what" archaeology in plain reading.
+
+This is the principle that prevents "superpowers providing specs over specs".
+
+## Enforcement (Play 2b)
+
+- Validator HARD-FAILs filenames matching `*-amendment-*.md`, `*-revision-*.md`, `*-update-*.md`, `*-followup-*.md` in `docs/superpowers/specs/`.
+- Validator WARNs on design-spec body diffs that add a `## Amendment` heading (allowed for ADRs only).
+- Validator WARNs on design-spec `status: delivered` without `delivered-by:` frontmatter.
+- Validator WARNs on `git diff` to decision-log lines older than 7 days (append-only convention).
+- This skill (`agent-context-quickstart`) loads this doc on session start so every subagent gets the rules cold.
+
+`.docs-discipline-allowlist` escape hatch for rare intentional exceptions; additions require decision-log entry citing rationale.
+
+## Origin story
+
+This discipline was codified after the **wedge-amendment incident (2026-05-17)**: a mid-flight design change to the wedge spec was captured as a separate `2026-05-16-improve-tab-amendment-design.md` file. A new session loaded the wedge spec, didn't see the amendment doc, went down the wrong path. The user had to intervene + correct.
+
+The lesson: convention alone doesn't prevent this — mechanical enforcement (Play 2b) does. The original Play 2b proposed "addendum threads on all living docs" — also wrong-shape because it conflated design specs (current-state SoT) with ADRs (point-in-time records). Different doc types need different patterns.
+
+Related: [[feedback_design_specs_edit_in_place]], [[project_adr_amendment_convention]], [[project_eight_purpose_taxonomy]]
diff --git a/docs/agent-context/onboarding-quick-start.md b/docs/agent-context/onboarding-quick-start.md
new file mode 100644
index 000000000..88c4d1f25
--- /dev/null
+++ b/docs/agent-context/onboarding-quick-start.md
@@ -0,0 +1,92 @@
+---
+title: 'VariScout Agent Onboarding Quick Start'
+purpose: agent-context
+tier: living
+audience: agent
+status: active
+topic: [ax, onboarding]
+last-verified: 2026-05-16
+---
+
+# VariScout Agent Onboarding Quick Start
+
+5-minute orientation for fresh subagents. Read this before touching any code or docs.
+
+---
+
+## What is VariScout?
+
+VariScout is a **browser-based structured investigation tool for process improvement specialists**. Customer data stays in the customer's Azure tenant (browser-only processing, no server-side compute). The deterministic stats engine is the authority; CoScout (AI) adds context, never recomputes stats. See [positioning.md](../01-vision/positioning.md) for the canonical one-liner and [ADR-082](../07-decisions/adr-082-wedge-architecture.md) for the V1 architectural record.
+
+## Current Strategic Direction (wedge pivot 2026-05-16)
+
+Single-product tool, one persona (improvement specialist), one SKU (€99/mo Azure tenant-wide).
+
+**6-tab workflow nav**: `Home · Projects · Process · Analyze · Investigation · Report`
+
+- "Improve" is a **stage inside Projects detail** (not a separate tab)
+- 4 stages: Charter → Approach → Improve → Sustainment
+- Hub = a single process investigation unit
+- Project-membership ACLs; no cross-AD-tenant invites
+- VariScout Process (enterprise, multi-Hub portfolio) is named-future only
+
+Canonical sources: [wedge spec](../superpowers/specs/2026-05-16-wedge-architecture-design.md) + [ADR-082](../07-decisions/adr-082-wedge-architecture.md). Many older docs predate this pivot — treat as historical until Phase C audit.
+
+---
+
+## Where to Look
+
+| Need                          | Go to                                                                   |
+| ----------------------------- | ----------------------------------------------------------------------- |
+| Strategic direction           | `docs/superpowers/specs/2026-05-16-wedge-architecture-design.md`        |
+| Architectural decisions (why) | `docs/07-decisions/` (live ADRs) · `docs/archive/adrs/` (superseded)    |
+| Product design (what)         | `docs/superpowers/specs/` (active) · `docs/archive/specs/` (historical) |
+| Current open decisions        | `docs/decision-log.md` — check here before re-opening any topic         |
+| Product overview              | `docs/OVERVIEW.md` · `docs/USER-JOURNEYS.md` · `docs/DATA-FLOW.md`      |
+| Package-specific context      | `packages/<pkg>/CLAUDE.md` · `apps/<app>/CLAUDE.md`                     |
+| Which package to edit         | See `package-router` skill                                              |
+| Store state glossary          | See `store-state-glossary` skill                                        |
+| Agent manifest / entry map    | `docs/llms.txt`                                                         |
+| Code smells / investigations  | `docs/investigations.md`                                                |
+
+**Package dependency direction**: `core → hooks → ui → apps`. Never import upward.
+
+**Sub-path exports**: updating a sub-path in `@variscout/core` (or any package) requires BOTH `package.json#exports` AND `tsconfig.json#paths` updated together.
+
+---
+
+## Hard Invariants
+
+Violating any of these is a hard error. ESLint enforces the language ones. Full index with canonical homes + enforcement mechanisms: [`.claude/INVARIANTS.md`](../../.claude/INVARIANTS.md).
+
+1. **Browser-only processing** — data never leaves the customer's tenant. No server-side aggregation, no external API calls with row data (ADR-059).
+2. **6 Zustand stores across 3 layers** — Document (×3): `useProjectStore`, `useInvestigationStore`, `useCanvasStore`; Annotation (×2): `useCanvasViewportStore`, `usePreferencesStore`; extra Annotation: `useActiveIPStore`; View (×1): `useViewStore`. No DataContext. Authoritative table: `packages/stores/CLAUDE.md`.
+3. **No statistical roll-up across heterogeneous units** — distributions, not aggregates; no Cpk arithmetic across different spec limits (ADR-073).
+4. **Language**: never write "root cause" — use "contribution" / "suspected cause" / "mechanism". Never call interactions "moderator/primary" — use `'ordinal'` / `'disordinal'`. ESLint rules enforce both.
+
+---
+
+## Common Pitfalls
+
+These cause bugs, test failures, or ESLint errors. Check before committing.
+
+- **`Math.random()` in any code or test** — forbidden everywhere. Use a seeded deterministic PRNG in tests; avoid randomness in production code entirely.
+- **Hardcoded hex colors in charts** — use `chartColors` / `chromeColors` from `@variscout/charts/colors`. No hex literals.
+- **`text-green-400` (or red/amber-400) without a paired `text-green-700`** — fails light-mode contrast. Always pair.
+- **`--no-verify` on commits** — never. Hooks enforce invariants; bypassing them hides real violations.
+- **"root cause" language** — ESLint rule `no-root-cause-language` enforces in prompts and comments. Use "contribution".
+- **Importing Dexie directly in domain stores** — only `apps/*/src/persistence/` and `apps/*/src/db/` may import Dexie. ESLint P7.2 enforces. R12 exception: `packages/stores/src/canvasViewportStore.ts`.
+- **Missing `@source` in `apps/*/src/index.css`** — Tailwind v4 silently breaks responsive utilities for any shared package missing an `@source` directive.
+- **Bare `useStore()` calls** — always use selectors: `useProjectStore(s => s.field)`.
+- **Single `package.json#exports` update** — sub-path exports need `tsconfig.json#paths` updated too, and vice versa.
+
+---
+
+## Skills to Reach For
+
+| Situation                                                       | Skill                                   |
+| --------------------------------------------------------------- | --------------------------------------- |
+| Need to know which package/CLAUDE.md to read for your work area | `package-router`                        |
+| Need to know what state lives in which store                    | `store-state-glossary`                  |
+| Writing or modifying tests                                      | `writing-tests`                         |
+| Need to find/query VariScout docs                               | `docs-toolbox` (available after Play 2) |
diff --git a/docs/agent-context/package-router.md b/docs/agent-context/package-router.md
new file mode 100644
index 000000000..9666699f9
--- /dev/null
+++ b/docs/agent-context/package-router.md
@@ -0,0 +1,72 @@
+---
+title: 'VariScout Package Router'
+purpose: agent-context
+tier: living
+audience: agent
+status: active
+topic: [ax, routing]
+last-verified: 2026-05-16
+---
+
+# VariScout Package Router
+
+Routing table: work area → primary CLAUDE.md → related context → ADRs → rules/invariants → suggested skills.
+
+Load the **Primary CLAUDE.md** first. Then load related ones as needed. Always check `docs/decision-log.md` before reopening any decided question.
+
+Package dependency direction (invariant): `core → hooks → ui → apps`. Never import upward.
+
+---
+
+## Routing Table
+
+| Work Area                                  | Primary CLAUDE.md                                                              | Related CLAUDE.mds                                       | Relevant ADRs                      | Rules/Invariants                                                                                                                                           | Suggested Skills       |
+| ------------------------------------------ | ------------------------------------------------------------------------------ | -------------------------------------------------------- | ---------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------- |
+| **Canvas / FRAME / Process Hub**           | `packages/ui/CLAUDE.md`                                                        | `packages/stores/CLAUDE.md`, `apps/azure/CLAUDE.md`      | ADR-070, ADR-076, ADR-078, ADR-082 | Canvas viewport = state not mechanism (8f); `Canvas` is canonical, `LayeredProcessViewWithCapability` deprecated                                           | `store-state-glossary` |
+| **Stats engine / math**                    | `packages/core/CLAUDE.md`                                                      | —                                                        | ADR-067, ADR-069, ADR-073          | `safeMath.ts` required; return `undefined` not NaN; never `Math.random()`; never `.toFixed()` on exported stat values                                      | `writing-tests`        |
+| **Chart components**                       | `packages/charts/CLAUDE.md`                                                    | `packages/core/CLAUDE.md`                                | ADR-002, ADR-005, ADR-051          | `.claude/invariants/charts.md`; `chartColors` only; `useChartTheme()`; pair `*-400` with `*-700` for labels                                                | —                      |
+| **Zustand stores**                         | `packages/stores/CLAUDE.md`                                                    | `apps/pwa/CLAUDE.md`, `apps/azure/CLAUDE.md`             | ADR-041, ADR-064, ADR-078, ADR-080 | Selector required; no bare `useStore()`; 3-layer boundary enforced by `layerBoundary.test.ts`                                                              | `store-state-glossary` |
+| **i18n strings / locale**                  | `packages/core/CLAUDE.md`                                                      | —                                                        | ADR-025                            | `.claude/invariants/i18n.md`; `registerLocaleLoaders()` before `preloadLocale()`; `formatStatistic()` not `.toFixed()`; `zh-Hans` → `zhHans.ts`            | `writing-tests`        |
+| **Azure-specific (storage / auth / sync)** | `apps/azure/CLAUDE.md`                                                         | `packages/stores/CLAUDE.md`                              | ADR-059, ADR-079                   | EasyAuth only (no MSAL); no PII in App Insights; ETag concurrency on hub-blob writes; R13 allow-list for sustainment direct writes                         | `store-state-glossary` |
+| **PWA-specific (persistence / Dexie)**     | `apps/pwa/CLAUDE.md`                                                           | `packages/stores/CLAUDE.md`                              | ADR-004, ADR-012, ADR-033, ADR-078 | Session-only by default; opt-in IndexedDB (single Hub-of-one); no cloud sync; no AI in free tier; `@source` directives required                            | `store-state-glossary` |
+| **UI primitives / components**             | `packages/ui/CLAUDE.md`                                                        | `packages/charts/CLAUDE.md`                              | ADR-045, ADR-056                   | `*Base` = primitive; `*WrapperBase` = app composition; semantic Tailwind only; no nested `<button>` in `<button>`                                          | —                      |
+| **React hooks**                            | `packages/hooks/CLAUDE.md`                                                     | `packages/core/CLAUDE.md`, `packages/stores/CLAUDE.md`   | ADR-041, ADR-045                   | Hooks consume stores, never define them; selectors not `getState()` in render paths; `use` prefix required                                                 | `writing-tests`        |
+| **CoScout prompts / AI**                   | `docs/01-vision/coscout-ax-design.md` (canonical entry point)                  | `packages/core/CLAUDE.md`                                | ADR-057, ADR-060, ADR-068, ADR-069 | `.claude/invariants/coscout-prompts.md`; `assembleCoScoutPrompt()` is entry point; tier 1 is cache-invariant; REF markers not raw data; never "root cause" | —                      |
+| **Response paths**                         | `docs/superpowers/specs/2026-04-27-process-learning-operating-model-design.md` | `packages/stores/CLAUDE.md`, `packages/ui/CLAUDE.md`     | ADR-080                            | 3 canvas response paths (V1 wedge); Sustainment auto-fires on `confirmed` + `implemented`                                                                  | `store-state-glossary` |
+| **Capability mode (Cpk)**                  | `packages/core/CLAUDE.md`                                                      | `packages/charts/CLAUDE.md`, `packages/ui/CLAUDE.md`     | ADR-047, ADR-074                   | `resolveMode()` produces `'capability'` from `standardIChartMetric`; never persist as `AnalysisMode`; Cpk grades target-relative to user-set target        | `writing-tests`        |
+| **Performance mode (multi-channel)**       | `packages/core/CLAUDE.md`                                                      | `packages/charts/CLAUDE.md`                              | ADR-047                            | `analysisMode === 'performance'`; `isPerformanceMode` removed — never re-introduce                                                                         | —                      |
+| **Yamazumi mode (cycle-time)**             | `packages/core/CLAUDE.md`                                                      | `packages/charts/CLAUDE.md`                              | ADR-047                            | `computeYamazumiData()` runs BEFORE stats; never call stats engine on raw event-log data                                                                   | —                      |
+| **Defect mode (events-to-rates)**          | `packages/core/CLAUDE.md`                                                      | `packages/charts/CLAUDE.md`                              | ADR-047                            | `computeDefectRates()` runs BEFORE stats; `PARETO_MAX_CATEGORIES=20` with "Others" aggregation                                                             | —                      |
+| **Process-flow / bottleneck mode**         | `packages/core/CLAUDE.md`                                                      | `packages/charts/CLAUDE.md`                              | ADR-074                            | `computeOutputRate`, `computeBottleneck` from `@variscout/core/throughput`                                                                                 | —                      |
+| **Projects tab / project membership**      | `docs/superpowers/specs/2026-05-16-wedge-architecture-design.md`               | `packages/stores/CLAUDE.md`, `apps/azure/CLAUDE.md`      | ADR-082                            | 4 stages: Charter→Approach→Improve→Sustainment; Lead/Member/Sponsor roles; project-membership ACLs                                                         | `store-state-glossary` |
+| **Investigation Wall / findings**          | `packages/core/CLAUDE.md`                                                      | `packages/stores/CLAUDE.md`, `packages/charts/CLAUDE.md` | ADR-064, ADR-073                   | `FindingSource` is discriminated union (6 variants, `chart` discriminant); always narrow before accessing; no cross-investigation aggregation              | `store-state-glossary` |
+| **Evidence Map**                           | `packages/charts/CLAUDE.md`                                                    | `packages/hooks/CLAUDE.md`, `packages/ui/CLAUDE.md`      | ADR-074, ADR-077                   | `EvidenceMapBase` is props-only; Layer 2+3 are Azure-only; click vs right-click contracts must not swap                                                    | —                      |
+| **Sample data**                            | `packages/data/CLAUDE.md`                                                      | `packages/core/CLAUDE.md`                                | —                                  | Data only, no logic; all exports from `src/index.ts`                                                                                                       | —                      |
+
+---
+
+## Quick Reference: Package Names
+
+| Import path            | Package             | Location           |
+| ---------------------- | ------------------- | ------------------ |
+| `@variscout/core`      | Core domain + stats | `packages/core/`   |
+| `@variscout/hooks`     | React hooks         | `packages/hooks/`  |
+| `@variscout/ui`        | UI components       | `packages/ui/`     |
+| `@variscout/charts`    | Chart components    | `packages/charts/` |
+| `@variscout/stores`    | Zustand stores      | `packages/stores/` |
+| `@variscout/data`      | Sample datasets     | `packages/data/`   |
+| `@variscout/pwa`       | Free PWA app        | `apps/pwa/`        |
+| `@variscout/azure-app` | Azure paid app      | `apps/azure/`      |
+
+---
+
+## How to Use This Table
+
+1. Identify your work area from the left column.
+2. Read the **Primary CLAUDE.md** in full.
+3. Skim **Related CLAUDE.mds** for boundary rules.
+4. Look up cited **ADRs** in `docs/07-decisions/` when the rationale matters.
+5. Check `.claude/invariants/` for machine-enforced constraints in the **Rules/Invariants** column. See `.claude/INVARIANTS.md` for the synthesizing index.
+6. Invoke the suggested skill if listed.
+
+If your work area isn't in the table, start at `CLAUDE.md` (root) and `docs/llms.txt`.
diff --git a/docs/agent-context/store-state-glossary.md b/docs/agent-context/store-state-glossary.md
new file mode 100644
index 000000000..69fc012fc
--- /dev/null
+++ b/docs/agent-context/store-state-glossary.md
@@ -0,0 +1,215 @@
+---
+title: 'VariScout Store State Glossary'
+purpose: agent-context
+tier: living
+audience: agent
+status: active
+topic: [ax, stores]
+last-verified: 2026-05-16
+verified-against-commit: 5ee58dc9
+---
+
+# VariScout Store State Glossary
+
+7 Zustand stores across 3 layers (ADR-078 + F4, shipped 2026-05-07). Read this before reading or mutating any store state.
+
+**Authoritative source**: `packages/stores/CLAUDE.md` + `packages/stores/src/*.ts`.
+
+**Test command**: `pnpm --filter @variscout/stores test` — `__tests__/layerBoundary.test.ts` enforces middleware presence/absence per layer.
+
+---
+
+## Layer Model
+
+| Layer               | Role                                                                | Portability test                       | Middleware                                               |
+| ------------------- | ------------------------------------------------------------------- | -------------------------------------- | -------------------------------------------------------- |
+| **Document** (×3)   | Portable project data — another analyst importing this hub needs it | `Yes → Document`                       | persist middleware (hub repository dispatch)             |
+| **Annotation** (×3) | Survives reload but not portable; scoped per-hub or per-user        | `Reload yes, portable no → Annotation` | persist middleware (Dexie or idb-keyval or localStorage) |
+| **View** (×1)       | Transient — does not survive reload                                 | `No → View`                            | no persist middleware                                    |
+
+**ADR-074 boundary**: cross-investigation aggregation is forbidden. `investigationStore` never aggregates across multiple hubs.
+
+---
+
+## Document Layer (3 stores)
+
+### `useProjectStore`
+
+**Location**: `packages/stores/src/projectStore.ts`
+**Persistence**: consumer-side serialisation via `useProjectActions` → `.vrs` file + Dexie (Azure)
+
+Primary store for the active hub's dataset + configuration. Clearing or loading a project resets the View layer via `useViewStore.clearTransientSelections()`.
+
+| Field               | Type                         | Description                                                                                               |
+| ------------------- | ---------------------------- | --------------------------------------------------------------------------------------------------------- |
+| `rawData`           | `DataRow[]`                  | All parsed rows from the uploaded CSV/paste                                                               |
+| `outcome`           | `string \| null`             | Selected Y outcome column name                                                                            |
+| `factors`           | `string[]`                   | Selected X factor column names                                                                            |
+| `specs`             | `SpecLimits`                 | USL/LSL/target for the primary outcome                                                                    |
+| `measureSpecs`      | `Record<string, SpecLimits>` | Per-column spec limits for multi-channel mode                                                             |
+| `analysisMode`      | `AnalysisMode`               | `'standard' \| 'performance' \| 'yamazumi' \| 'defect'` — never `'capability'` (that is a `ResolvedMode`) |
+| `projectId`         | `string`                     | UUID for the current hub                                                                                  |
+| `projectName`       | `string`                     | Display name                                                                                              |
+| `dataFilename`      | `string \| null`             | Original file name for display                                                                            |
+| `subgroupConfig`    | `SubgroupConfig`             | n for subgrouped I-chart / Xbar-R; default n=5+ for capability                                            |
+| `filters`           | `FilterAction[]`             | Active row-level filters                                                                                  |
+| `stageColumn`       | `string \| null`             | Column used as categorical stage/group axis                                                               |
+| `timeColumn`        | `string \| null`             | Column for chronological ordering                                                                         |
+| `dataQualityReport` | `DataQualityReport \| null`  | Parse-time quality report (blank rows, type mismatches)                                                   |
+
+**Key actions**: `loadProject(SerializedProject)`, `newProject()`, `setOutcome()`, `setFactors()`, `setMeasureSpec()`, `addFilter()`, `removeFilter()`.
+
+Note: `isPerformanceMode` was deleted in F4 — never re-introduce. Use `analysisMode === 'performance'` or `getStrategy()` from `@variscout/core`.
+
+---
+
+### `useInvestigationStore`
+
+**Location**: `packages/stores/src/investigationStore.ts`
+**Persistence**: session-only today; future `HubRepository.dispatch`
+
+Owns all investigation-domain entities for the active hub. Never aggregates across hubs (ADR-073).
+
+| Field                     | Type                      | Description                                                                                                                                                             |
+| ------------------------- | ------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `findings`                | `Finding[]`               | Pinned observations (chart + table + journal types); `FindingSource` is a discriminated union with `chart` discriminant — always narrow before accessing variant fields |
+| `questions`               | `Question[]`              | Investigation questions (tree, max depth 3, max 8 children); tree traversal via `wouldCreateCycle()`                                                                    |
+| `hypotheses`              | `Hypothesis[]`            | `SuspectedCause` entities (first-class per ADR-064); `selectedForImprovement: true` → triggers HMW brainstorm                                                           |
+| `causalLinks`             | `CausalLink[]`            | Directed edges in the causal graph; cycle prevention is mandatory — `addCausalLink` calls `wouldCreateCycle()` internally, never bypass                                 |
+| `categories`              | `InvestigationCategory[]` | User-defined groupings for findings/questions                                                                                                                           |
+| `problemContributionTree` | `GateNode \| undefined`   | AND-gate tree for the Investigation Wall; highlights live in `useViewStore`                                                                                             |
+
+**Legacy note**: `causeRole: 'primary' | 'contributing'` is deprecated. Multiple hubs can coexist; each `selectedForImprovement: true` hub triggers one HMW brainstorm.
+
+**Persistence**: `findings`, `questions`, `hypotheses`, `causalLinks`, `categories` serialize via `useProjectActions` into `.vrs`. New entities need both `apps/azure/src/db/schema.ts` and `useEditorDataFlow.ts` updates.
+
+---
+
+### `useCanvasStore`
+
+**Location**: `packages/stores/src/canvasStore.ts`
+**Persistence**: `dispatch(CanvasAction)` + hub repository
+
+Owns the FRAME / Process Hub canvas document. The canvas is the substrate for the Process tab.
+
+| Field                    | Type                   | Description                                                    |
+| ------------------------ | ---------------------- | -------------------------------------------------------------- |
+| `canonicalMap`           | `ProcessMap`           | The process map (nodes = steps, edges = connections)           |
+| `outcomes`               | `OutcomeSpec[]`        | Outcome specs attached to the hub (from `ProcessHub.outcomes`) |
+| `primaryScopeDimensions` | `string[]`             | Primary scope dimensions for the hub                           |
+| `canonicalMapVersion`    | `string`               | Version string for optimistic concurrency                      |
+| `undoStack`              | `CanvasHistoryEntry[]` | History for undo (cap 50)                                      |
+| `redoStack`              | `CanvasHistoryEntry[]` | History for redo (cap 50)                                      |
+
+**Key actions**: `dispatch(CanvasAction)`, `addStep()`, `removeStep()`, `connectSteps()`, `undo()`, `redo()`.
+
+---
+
+## Annotation Layer (3 stores)
+
+### `useCanvasViewportStore`
+
+**Location**: `packages/stores/src/canvasViewportStore.ts`
+**Persistence**: Dexie DB `variscout-canvas-viewport` (R12 ESLint exception — this store imports Dexie directly for cross-app UI state)
+**`STORE_LAYER`**: `'annotation-per-hub'`
+
+Per-hub canvas viewport state. Call `rehydrateCanvasViewport(hubId)` on hub open; debounced `persistCanvasViewport(hubId)` on mutation.
+
+| Field              | Type                                           | Description                                               |
+| ------------------ | ---------------------------------------------- | --------------------------------------------------------- |
+| `viewMode`         | `'map' \| 'wall'`                              | Whether showing the process map or the Investigation Wall |
+| `viewports`        | `Record<ProcessHubId, CanvasViewportSnapshot>` | Per-hub zoom + pan + level state                          |
+| `selection`        | `Set<NodeId>`                                  | Currently selected canvas nodes                           |
+| `chartClusters`    | `ChartClusterState[]`                          | Positioned chart clusters on the Wall                     |
+| `andCheckSnapshot` | `AndCheckSnapshot \| null`                     | Snapshot of the AND-gate hold/total for the Wall          |
+| `pendingComments`  | `PendingComment[]`                             | Optimistic comment queue                                  |
+
+**Canvas viewport principle (8f)**: viewport state is a state concept, not a rendering mechanism. Pan/zoom + active level/lens are unified here; each renderer (SVG/DOM) paints in its native medium.
+
+---
+
+### `usePreferencesStore`
+
+**Location**: `packages/stores/src/preferencesStore.ts`
+**Persistence**: idb-keyval, key `'variscout-preferences'`
+**`STORE_LAYER`**: `'annotation-per-user'`
+
+Durable per-user preferences that survive reload but are not portable to other analysts.
+
+| Field                  | Type             | Description                                                                                                   |
+| ---------------------- | ---------------- | ------------------------------------------------------------------------------------------------------------- |
+| `activeView`           | `WorkspaceView`  | Current workspace tab: `'dashboard' \| 'frame' \| 'analysis' \| 'investigation' \| 'improvement' \| 'report'` |
+| `piActiveTab`          | `PITab`          | Active PI panel tab: `'stats' \| 'questions' \| 'journal'`                                                    |
+| `isPISidebarOpen`      | `boolean`        | Whether the PI sidebar is open                                                                                |
+| `isCoScoutOpen`        | `boolean`        | Whether the CoScout panel is open (Azure only)                                                                |
+| `aiEnabled`            | `boolean`        | User preference for AI features                                                                               |
+| `timeLens`             | `TimeLens`       | Active time lens selection (default `DEFAULT_TIME_LENS`)                                                      |
+| `riskAxisConfig`       | `RiskAxisConfig` | Improvement matrix risk axis configuration                                                                    |
+| `budgetConfig`         | `BudgetConfig`   | Improvement budget configuration                                                                              |
+| `isIPTeamRailExpanded` | `boolean`        | Whether the IP team rail is expanded                                                                          |
+
+**Deleted stores**: `useSessionStore` and `useImprovementStore` were deleted in F4 (2026-05-07). Never re-introduce them. Preferences → `usePreferencesStore`; transient view state → `useViewStore`; improvement plan state → `usePreferencesStore` (relocated in F4).
+
+---
+
+### `useActiveIPStore`
+
+**Location**: `packages/stores/src/activeIPStore.ts`
+**Persistence**: localStorage, key `variscout:activeIP:{hubId}:{userId}` (with URL-encoded scope parts)
+**`STORE_LAYER`**: `'annotation-per-user'`
+
+Tracks the active improvement project per hub per user. Scoped to prevent cross-user bleed.
+
+| Field       | Type                            | Description                            |
+| ----------- | ------------------------------- | -------------------------------------- |
+| `activeIPs` | `Record<string, ActiveIPState>` | Map of storage key → `{ ipId, setAt }` |
+
+**Key actions**: `getActiveIP(scope)`, `setActiveIP(scope, ipId)`, `clearActiveIP(scope)`, `rehydrateActiveIP(scope)`.
+
+---
+
+## View Layer (1 store)
+
+### `useViewStore`
+
+**Location**: `packages/stores/src/viewStore.ts`
+**Persistence**: NONE — transient, cleared on `projectStore.loadProject()` / `newProject()`
+**`STORE_LAYER`**: `'view'`
+
+All state that does not survive browser reload. Test pattern: `beforeEach(() => useViewStore.setState(useViewStore.getInitialState()))`.
+
+| Field                          | Type                         | Description                                            |
+| ------------------------------ | ---------------------------- | ------------------------------------------------------ |
+| `highlightRowIndex`            | `number \| null`             | Row highlighted in data table (from chart point click) |
+| `highlightedChartPoint`        | `number \| null`             | Chart point highlighted (from data table row click)    |
+| `highlightedFindingId`         | `string \| null`             | Finding ID highlighted for bidirectional navigation    |
+| `expandedQuestionId`           | `string \| null`             | Question expanded in PI panel questions tab            |
+| `pendingChartFocus`            | `string \| null`             | Set by `navigate_to` CoScout tool; consumed by Editor  |
+| `piOverflowView`               | `'data' \| 'whatif' \| null` | Secondary overflow view in PI panel                    |
+| `isDataTableOpen`              | `boolean`                    | Data table modal open state                            |
+| `focusedQuestionId`            | `string \| null`             | Question focused on the Investigation Wall             |
+| `highlightedImprovementIdeaId` | `string \| null`             | Idea highlighted in prioritization matrix              |
+| `improvementActiveView`        | `'plan' \| 'track'`          | Active tab in the IMPROVE workspace                    |
+| `selectedPoints`               | `Set<number>`                | Multi-point brushing selection (Minitab-style)         |
+| `selectionIndexMap`            | `Map<number, number>`        | Row index → display point index mapping                |
+
+Note: `problemContributionTree` highlights live in this store; the tree itself lives in `useInvestigationStore`.
+
+---
+
+## Usage Rules (enforced by ESLint + tests)
+
+- **Selectors required**: `useProjectStore(s => s.field)` — never bare `useStore()`.
+- **No DataContext** — all shared state via stores (ADR-041).
+- **Document stores never import `dexie` directly** (ESLint P7.2). Only `apps/*/src/persistence/**` and `apps/*/src/db/**` are whitelisted. R12 exception: `canvasViewportStore.ts`.
+- **Cross-store imperative action calls are allowed**; reactive subscriptions between stores are forbidden.
+- **Sustainment direct writes**: `sustainmentRecords`, `sustainmentReviews`, `controlHandoffs` are NOT yet hub-domain-dispatched — direct `apps/azure/src/services/localDb.ts` writes (R13 allow-listed). F5 may unify.
+
+---
+
+## Related
+
+- `packages/stores/CLAUDE.md` — authoritative one-page summary with layer table
+- `packages/stores/src/__tests__/layerBoundary.test.ts` — enforces middleware presence/absence per layer
+- ADR-041, ADR-064, ADR-065, ADR-078, ADR-080
+- `docs/superpowers/specs/2026-05-07-data-flow-foundation-f4-three-layer-state-design.md` — F4 three-layer design spec
diff --git a/docs/archive/README.md b/docs/archive/README.md
index 2befadb58..e887fba06 100644
--- a/docs/archive/README.md
+++ b/docs/archive/README.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: 'Archive'
-audience: [engineer, architect]
+audience: human
 category: reference
-status: reference
+status: active
 related: [doc-prune-audit, agent-docs-architecture]
 last-reviewed: 2026-04-17
 ---
diff --git a/docs/archive/adrs/adr-006-edition-system.md b/docs/archive/adrs/adr-006-edition-system.md
index 7240c87db..0d55b61f7 100644
--- a/docs/archive/adrs/adr-006-edition-system.md
+++ b/docs/archive/adrs/adr-006-edition-system.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-006: Edition System'
 ---
 
diff --git a/docs/archive/adrs/adr-016-teams-integration.md b/docs/archive/adrs/adr-016-teams-integration.md
index c641d5563..8cc31d27f 100644
--- a/docs/archive/adrs/adr-016-teams-integration.md
+++ b/docs/archive/adrs/adr-016-teams-integration.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-016: Microsoft Teams Integration'
 ---
 
diff --git a/docs/archive/adrs/adr-018-channel-mention-workflow.md b/docs/archive/adrs/adr-018-channel-mention-workflow.md
index 36b133dac..47b0e6ba7 100644
--- a/docs/archive/adrs/adr-018-channel-mention-workflow.md
+++ b/docs/archive/adrs/adr-018-channel-mention-workflow.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-018: Channel @Mention Workflow — Mobile Finding to Team Action'
 ---
 
diff --git a/docs/archive/adrs/adr-021-security-evaluation.md b/docs/archive/adrs/adr-021-security-evaluation.md
index c422daaf6..e0b1aff99 100644
--- a/docs/archive/adrs/adr-021-security-evaluation.md
+++ b/docs/archive/adrs/adr-021-security-evaluation.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-021 Security Evaluation: Teams Integration'
 ---
 
diff --git a/docs/archive/adrs/adr-022-knowledge-layer-architecture.md b/docs/archive/adrs/adr-022-knowledge-layer-architecture.md
index caa4d11c3..bd3f664fb 100644
--- a/docs/archive/adrs/adr-022-knowledge-layer-architecture.md
+++ b/docs/archive/adrs/adr-022-knowledge-layer-architecture.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-022: Knowledge Layer Architecture'
 ---
 
diff --git a/docs/archive/adrs/adr-024-scouting-report.md b/docs/archive/adrs/adr-024-scouting-report.md
index af44dc750..063ea882e 100644
--- a/docs/archive/adrs/adr-024-scouting-report.md
+++ b/docs/archive/adrs/adr-024-scouting-report.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-024: Scouting Report — Dynamic Analysis Report View'
-audience: [analyst, engineer]
+audience: human
 category: architecture
 status: superseded
 superseded-by: adr-037
diff --git a/docs/archive/adrs/adr-026-knowledge-base-sharepoint-first.md b/docs/archive/adrs/adr-026-knowledge-base-sharepoint-first.md
index 455791fe4..53bbe37c9 100644
--- a/docs/archive/adrs/adr-026-knowledge-base-sharepoint-first.md
+++ b/docs/archive/adrs/adr-026-knowledge-base-sharepoint-first.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-026: Knowledge Base Strategy — SharePoint-First with Remote Retrieval'
 ---
 
diff --git a/docs/archive/adrs/adr-030-unified-file-picker.md b/docs/archive/adrs/adr-030-unified-file-picker.md
index ae2b0023f..79df662f3 100644
--- a/docs/archive/adrs/adr-030-unified-file-picker.md
+++ b/docs/archive/adrs/adr-030-unified-file-picker.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-030: Unified File Picker (OneDrive v8)'
 status: superseded
 date: 2026-03-19
diff --git a/docs/archive/adrs/adr-043-teams-entry-experience.md b/docs/archive/adrs/adr-043-teams-entry-experience.md
index a0a9cd47e..8fe793627 100644
--- a/docs/archive/adrs/adr-043-teams-entry-experience.md
+++ b/docs/archive/adrs/adr-043-teams-entry-experience.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-043: Teams Entry Experience'
 status: superseded
 date: 2026-03-22
diff --git a/docs/archive/adrs/adr-064-suspected-cause-hub-model.md b/docs/archive/adrs/adr-064-suspected-cause-hub-model.md
index 2de6f1119..44f602080 100644
--- a/docs/archive/adrs/adr-064-suspected-cause-hub-model.md
+++ b/docs/archive/adrs/adr-064-suspected-cause-hub-model.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'ADR-064: SuspectedCause Hub Model & Investigation Reframing'
 ---
 
diff --git a/docs/archive/plans/2026-04-26-phase-6-implementation-plan.md b/docs/archive/plans/2026-04-26-phase-6-implementation-plan.md
index 881c064a8..ce126eefa 100644
--- a/docs/archive/plans/2026-04-26-phase-6-implementation-plan.md
+++ b/docs/archive/plans/2026-04-26-phase-6-implementation-plan.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Phase 6 — Sustainment & Control Handoff Implementation Plan
-status: deferred
+status: superseded
 ---
 
 > **⚠️ Archived 2026-05-16** — superseded by the [wedge pivot](../../superpowers/specs/2026-05-16-wedge-architecture-design.md) + [ADR-082](../../07-decisions/adr-082-wedge-architecture.md). Sustainment + Handoff + Control Plan implementation replaced by [ADR-080](../../07-decisions/adr-080-sustainment-auto-fire-pattern.md) auto-fire pattern + wedge spec §3.2 Sustainment stage. Handoff folded into Sustainment closure.
diff --git a/docs/archive/specs/2026-03-16-ai-integration-evaluation.md b/docs/archive/specs/2026-03-16-ai-integration-evaluation.md
index 7baac99e9..4e039e256 100644
--- a/docs/archive/specs/2026-03-16-ai-integration-evaluation.md
+++ b/docs/archive/specs/2026-03-16-ai-integration-evaluation.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: 'AI Integration Quality Evaluation'
-status: delivered
+status: active
 ---
 
 # AI Integration Quality Evaluation — VariScout
diff --git a/docs/archive/specs/2026-03-16-code-review-design.md b/docs/archive/specs/2026-03-16-code-review-design.md
index 900048b6b..f98d5b964 100644
--- a/docs/archive/specs/2026-03-16-code-review-design.md
+++ b/docs/archive/specs/2026-03-16-code-review-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Systematic Code Review — VariScout Lite
 date: 2026-03-16
 type: code-review
-status: delivered
+status: active
 ---
 
 # Systematic Code Review — VariScout Lite
diff --git a/docs/archive/specs/2026-03-17-documentation-methodology-upgrade-design.md b/docs/archive/specs/2026-03-17-documentation-methodology-upgrade-design.md
index 17ebd03fc..be0737c6d 100644
--- a/docs/archive/specs/2026-03-17-documentation-methodology-upgrade-design.md
+++ b/docs/archive/specs/2026-03-17-documentation-methodology-upgrade-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Documentation Methodology Upgrade — Lessons from tblsum
-audience: [engineer, analyst]
+audience: human
 category: architecture
 status: draft
 date: 2026-03-17
diff --git a/docs/archive/specs/2026-03-17-navigation-architecture-design.md b/docs/archive/specs/2026-03-17-navigation-architecture-design.md
index 525984e8e..580f752bf 100644
--- a/docs/archive/specs/2026-03-17-navigation-architecture-design.md
+++ b/docs/archive/specs/2026-03-17-navigation-architecture-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Navigation Architecture + Invisible Phase Intelligence
-audience: [developer, designer]
+audience: human
 category: architecture
 status: draft
 related: [navigation, mobile, teams, accessibility, phase-detection, toolbar]
diff --git a/docs/archive/specs/2026-03-17-navigation-audit.md b/docs/archive/specs/2026-03-17-navigation-audit.md
index 00639169f..6a48eeb2b 100644
--- a/docs/archive/specs/2026-03-17-navigation-audit.md
+++ b/docs/archive/specs/2026-03-17-navigation-audit.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Navigation Audit — Findings & Action Items
-audience: [developer, designer]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [navigation, mobile, teams, accessibility, onboarding]
 date: 2026-03-17
 ---
diff --git a/docs/archive/specs/2026-03-17-platform-opportunities.md b/docs/archive/specs/2026-03-17-platform-opportunities.md
index 8eec9c5d2..2437ce540 100644
--- a/docs/archive/specs/2026-03-17-platform-opportunities.md
+++ b/docs/archive/specs/2026-03-17-platform-opportunities.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Azure / AI Foundry / Teams Platform Opportunities
-audience: [developer, product]
+audience: human
 category: strategy
 status: active
 related: [azure, ai, teams, platform]
diff --git a/docs/archive/specs/2026-03-17-teams-compliance-audit.md b/docs/archive/specs/2026-03-17-teams-compliance-audit.md
index 7cd88410b..a9fdcf535 100644
--- a/docs/archive/specs/2026-03-17-teams-compliance-audit.md
+++ b/docs/archive/specs/2026-03-17-teams-compliance-audit.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Teams App Store + Microsoft Patterns Compliance Audit
-audience: [developer]
+audience: human
 category: compliance
 status: archived
 related: [teams, accessibility, azure, navigation]
diff --git a/docs/archive/specs/2026-03-19-knowledge-base-folder-search-design.md b/docs/archive/specs/2026-03-19-knowledge-base-folder-search-design.md
index 314f65556..7be83ed95 100644
--- a/docs/archive/specs/2026-03-19-knowledge-base-folder-search-design.md
+++ b/docs/archive/specs/2026-03-19-knowledge-base-folder-search-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: 'Knowledge Base: Document Upload, Folder Selection & Permissions Design'
-audience: [engineer, analyst]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [knowledge-base, sharepoint, coscout, adr-026]
 ---
 
diff --git a/docs/archive/specs/2026-03-20-improvement-prioritization-design.md b/docs/archive/specs/2026-03-20-improvement-prioritization-design.md
index d01daec5f..d51896046 100644
--- a/docs/archive/specs/2026-03-20-improvement-prioritization-design.md
+++ b/docs/archive/specs/2026-03-20-improvement-prioritization-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Improvement Prioritization Design
-audience: [engineer, analyst]
+audience: human
 category: workflow
-status: delivered
+status: active
 related: [improve, prioritization, risk, timeframe, cost, matrix, ideas, actions, adr-035]
 ---
 
diff --git a/docs/archive/specs/2026-03-20-reporting-workspaces-design.md b/docs/archive/specs/2026-03-20-reporting-workspaces-design.md
index c9332e65e..63f4d83e4 100644
--- a/docs/archive/specs/2026-03-20-reporting-workspaces-design.md
+++ b/docs/archive/specs/2026-03-20-reporting-workspaces-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Reporting Workspaces Design
-audience: [engineer, analyst, manager]
+audience: human
 category: workflow
-status: delivered
+status: active
 related:
   [report, workspaces, analysis, findings, improvement, audience, two-voices, adr-024, adr-031]
 ---
diff --git a/docs/archive/specs/2026-03-20-yamazumi-analysis-mode-design.md b/docs/archive/specs/2026-03-20-yamazumi-analysis-mode-design.md
index fa22c8c9f..d6efb95e2 100644
--- a/docs/archive/specs/2026-03-20-yamazumi-analysis-mode-design.md
+++ b/docs/archive/specs/2026-03-20-yamazumi-analysis-mode-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Yamazumi Analysis Mode
-audience: [analyst, engineer]
+audience: human
 category: analysis
-status: delivered
+status: active
 related: [yamazumi, lean, time-study, waste, value-stream, takt-time, activity-type]
 ---
 
diff --git a/docs/archive/specs/2026-03-21-analysis-flow-design.md b/docs/archive/specs/2026-03-21-analysis-flow-design.md
index 7caa99470..a07e3225e 100644
--- a/docs/archive/specs/2026-03-21-analysis-flow-design.md
+++ b/docs/archive/specs/2026-03-21-analysis-flow-design.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: Analysis Flow Documentation + Capability Mode Features
 status: draft
 ---
diff --git a/docs/archive/specs/2026-03-21-capability-time-subgrouping.md b/docs/archive/specs/2026-03-21-capability-time-subgrouping.md
index cbd9a17f2..0b5c4c873 100644
--- a/docs/archive/specs/2026-03-21-capability-time-subgrouping.md
+++ b/docs/archive/specs/2026-03-21-capability-time-subgrouping.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'Capability Mode: Time-Based Subgrouping + Journey Integration'
 status: draft
 ---
diff --git a/docs/archive/specs/2026-03-21-yamazumi-reporting-design.md b/docs/archive/specs/2026-03-21-yamazumi-reporting-design.md
index 18dccd85b..2373701fa 100644
--- a/docs/archive/specs/2026-03-21-yamazumi-reporting-design.md
+++ b/docs/archive/specs/2026-03-21-yamazumi-reporting-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Yamazumi-Aware Reporting View
-audience: [engineer, analyst]
+audience: human
 category: workflow
 status: draft
 related: [yamazumi, reporting, lean, findings]
diff --git a/docs/archive/specs/2026-03-22-mobile-ux-improvements-design.md b/docs/archive/specs/2026-03-22-mobile-ux-improvements-design.md
index e2bc3a032..198393df9 100644
--- a/docs/archive/specs/2026-03-22-mobile-ux-improvements-design.md
+++ b/docs/archive/specs/2026-03-22-mobile-ux-improvements-design.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'Mobile UX Improvements — Design Spec'
 status: draft
 ---
diff --git a/docs/archive/specs/2026-03-22-mode-aware-reports-design.md b/docs/archive/specs/2026-03-22-mode-aware-reports-design.md
index c6dea1e8f..456de89e5 100644
--- a/docs/archive/specs/2026-03-22-mode-aware-reports-design.md
+++ b/docs/archive/specs/2026-03-22-mode-aware-reports-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Mode-Aware Reports
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: delivered
+status: active
 related: [report-view, analysis-mode, capability, performance, yamazumi, two-voices]
 ---
 
diff --git a/docs/archive/specs/2026-03-22-sharing-continuity-design.md b/docs/archive/specs/2026-03-22-sharing-continuity-design.md
index b4fdd0f53..e9be6684e 100644
--- a/docs/archive/specs/2026-03-22-sharing-continuity-design.md
+++ b/docs/archive/specs/2026-03-22-sharing-continuity-design.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'Sharing & Continuity UX Design'
 status: draft
 ---
diff --git a/docs/archive/specs/2026-03-22-teams-entry-experience-design.md b/docs/archive/specs/2026-03-22-teams-entry-experience-design.md
index 675cad713..c8faf3b9f 100644
--- a/docs/archive/specs/2026-03-22-teams-entry-experience-design.md
+++ b/docs/archive/specs/2026-03-22-teams-entry-experience-design.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'Teams Entry Experience Redesign'
 status: superseded
 ---
diff --git a/docs/archive/specs/2026-03-23-architecture-design-exploration.md b/docs/archive/specs/2026-03-23-architecture-design-exploration.md
index 456b37076..d61d75613 100644
--- a/docs/archive/specs/2026-03-23-architecture-design-exploration.md
+++ b/docs/archive/specs/2026-03-23-architecture-design-exploration.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: 'Architecture Design Exploration'
-audience: [developer]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [adr-045, ddd, feature-sliced-design, strategy-pattern, domain-events]
 ---
 
diff --git a/docs/archive/specs/2026-03-23-event-driven-architecture-design.md b/docs/archive/specs/2026-03-23-event-driven-architecture-design.md
index d68025f46..91f9f15c1 100644
--- a/docs/archive/specs/2026-03-23-event-driven-architecture-design.md
+++ b/docs/archive/specs/2026-03-23-event-driven-architecture-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Event-Driven Architecture Transition
-audience: [engineer]
+audience: human
 category: architecture
 status: superseded
 related: [domain-events, strategy-pattern, eslint-boundaries, zustand, mitt]
diff --git a/docs/archive/specs/2026-03-24-coscout-knowledge-catalyst-design.md b/docs/archive/specs/2026-03-24-coscout-knowledge-catalyst-design.md
index 70da173d5..735f807f6 100644
--- a/docs/archive/specs/2026-03-24-coscout-knowledge-catalyst-design.md
+++ b/docs/archive/specs/2026-03-24-coscout-knowledge-catalyst-design.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'Design: CoScout Knowledge Catalyst'
 status: draft
 ---
diff --git a/docs/archive/specs/2026-03-28-dashboard-chrome-redesign.md b/docs/archive/specs/2026-03-28-dashboard-chrome-redesign.md
index be55a38c7..f768a6312 100644
--- a/docs/archive/specs/2026-03-28-dashboard-chrome-redesign.md
+++ b/docs/archive/specs/2026-03-28-dashboard-chrome-redesign.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Dashboard Chrome Redesign
 date: 2026-03-28
-status: delivered
-audience: [developer, designer]
+status: active
+audience: human
 category: design-spec
 related: [header, toolbar, stats-panel, navigation, panels, layout, workspaces]
 ---
diff --git a/docs/archive/specs/2026-03-28-process-health-projection-toolbar-design.md b/docs/archive/specs/2026-03-28-process-health-projection-toolbar-design.md
index 4a6cd1dd7..98d6251d8 100644
--- a/docs/archive/specs/2026-03-28-process-health-projection-toolbar-design.md
+++ b/docs/archive/specs/2026-03-28-process-health-projection-toolbar-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Process Health & Projection Toolbar
 date: 2026-03-28
-status: delivered
-audience: [developer, designer]
+status: active
+audience: human
 category: design-spec
 related: [dashboard, toolbar, stats, variation, what-if, findings, improvement, projection]
 ---
diff --git a/docs/archive/specs/2026-03-28-process-intelligence-panel-design.md b/docs/archive/specs/2026-03-28-process-intelligence-panel-design.md
index 2e532f32b..0d79a578d 100644
--- a/docs/archive/specs/2026-03-28-process-intelligence-panel-design.md
+++ b/docs/archive/specs/2026-03-28-process-intelligence-panel-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Process Intelligence Panel
 date: 2026-03-28
 status: superseded
-audience: [developer, designer]
+audience: human
 category: design-spec
 related: [stats-panel, what-if, target-discovery, sidebar, process-health-bar]
 ---
diff --git a/docs/archive/specs/2026-03-29-adaptive-boxplot-categories-design.md b/docs/archive/specs/2026-03-29-adaptive-boxplot-categories-design.md
index dcbb0e6b2..222729184 100644
--- a/docs/archive/specs/2026-03-29-adaptive-boxplot-categories-design.md
+++ b/docs/archive/specs/2026-03-29-adaptive-boxplot-categories-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Adaptive Boxplot Category Limits
-audience: [developer, analyst]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [boxplot, charts, wide-form, responsive]
 ---
 
diff --git a/docs/archive/specs/2026-03-29-capability-mode-coherence-design.md b/docs/archive/specs/2026-03-29-capability-mode-coherence-design.md
index e94398304..e3e006625 100644
--- a/docs/archive/specs/2026-03-29-capability-mode-coherence-design.md
+++ b/docs/archive/specs/2026-03-29-capability-mode-coherence-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Capability Mode Coherence
-audience: [engineer]
+audience: human
 category: analysis
-status: delivered
+status: active
 related: [capability, cpk, boxplot, i-chart, process-health-bar]
 ---
 
diff --git a/docs/archive/specs/2026-03-29-display-density-design.md b/docs/archive/specs/2026-03-29-display-density-design.md
index c39da696d..4cbd7ba66 100644
--- a/docs/archive/specs/2026-03-29-display-density-design.md
+++ b/docs/archive/specs/2026-03-29-display-density-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Display Density — Global UI Scaling
-audience: [developer, designer]
+audience: human
 category: architecture
 status: draft
 related: [accessibility, typography, settings, density]
diff --git a/docs/archive/specs/2026-03-29-probability-plot-enhancement-design.md b/docs/archive/specs/2026-03-29-probability-plot-enhancement-design.md
index 8e9d272cc..9264ee3e4 100644
--- a/docs/archive/specs/2026-03-29-probability-plot-enhancement-design.md
+++ b/docs/archive/specs/2026-03-29-probability-plot-enhancement-design.md
@@ -1,4 +1,6 @@
 ---
+tier: stable
+purpose: remember
 title: 'Probability Plot Enhancement Design'
 status: draft
 ---
diff --git a/docs/archive/specs/2026-03-29-wide-form-stack-columns-design.md b/docs/archive/specs/2026-03-29-wide-form-stack-columns-design.md
index 0625b63ef..fd133e1e5 100644
--- a/docs/archive/specs/2026-03-29-wide-form-stack-columns-design.md
+++ b/docs/archive/specs/2026-03-29-wide-form-stack-columns-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Wide-Form Data Support — Stack Columns
-audience: [developer, analyst]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [column-mapping, parser, data-ingestion, performance-mode]
 ---
 
diff --git a/docs/archive/specs/2026-03-30-question-driven-eda-design.md b/docs/archive/specs/2026-03-30-question-driven-eda-design.md
index 933229bb3..6455b35c9 100644
--- a/docs/archive/specs/2026-03-30-question-driven-eda-design.md
+++ b/docs/archive/specs/2026-03-30-question-driven-eda-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Question-Driven EDA — CoScout Analysis Planning
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: delivered
+status: active
 related: [eda-mental-model, coscout, factor-intelligence, investigation, findings]
 ---
 
diff --git a/docs/archive/specs/2026-04-01-app-insights-telemetry-design.md b/docs/archive/specs/2026-04-01-app-insights-telemetry-design.md
index 819b4c4e1..3a698752c 100644
--- a/docs/archive/specs/2026-04-01-app-insights-telemetry-design.md
+++ b/docs/archive/specs/2026-04-01-app-insights-telemetry-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Application Insights Telemetry Design
-status: delivered
+status: active
 related: [app-insights, telemetry, security, observability]
 ---
 
diff --git a/docs/archive/specs/2026-04-01-data-view-consolidation-design.md b/docs/archive/specs/2026-04-01-data-view-consolidation-design.md
index 4cf2fd4ac..6381b1bff 100644
--- a/docs/archive/specs/2026-04-01-data-view-consolidation-design.md
+++ b/docs/archive/specs/2026-04-01-data-view-consolidation-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Data View Consolidation
-status: delivered
+status: active
 related: [data-table, process-intelligence-panel, dashboard-layout, performance]
 ---
 
diff --git a/docs/archive/specs/2026-04-01-header-redesign-design.md b/docs/archive/specs/2026-04-01-header-redesign-design.md
index fe8a75592..75cd752c0 100644
--- a/docs/archive/specs/2026-04-01-header-redesign-design.md
+++ b/docs/archive/specs/2026-04-01-header-redesign-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Header Redesign — Merged Toolbar + Workspace Tabs
 date: 2026-04-01
 status: superseded
-audience: [developer, designer]
+audience: human
 category: design-spec
 related: [header, toolbar, workspace-tabs, auto-save, report-workspace, navigation, panels]
 ---
diff --git a/docs/archive/specs/2026-04-01-process-intelligence-panel-redesign.md b/docs/archive/specs/2026-04-01-process-intelligence-panel-redesign.md
index f4f6e6719..b73c59a38 100644
--- a/docs/archive/specs/2026-04-01-process-intelligence-panel-redesign.md
+++ b/docs/archive/specs/2026-04-01-process-intelligence-panel-redesign.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Process Intelligence Panel Redesign
 date: 2026-04-01
-status: delivered
-audience: [developer, designer]
+status: active
+audience: human
 category: design-spec
 related: [stats-panel, questions, journal, findings, investigation, pi-panel, adr-056]
 ---
diff --git a/docs/archive/specs/2026-04-02-improvement-hub-design.md b/docs/archive/specs/2026-04-02-improvement-hub-design.md
index aea00a216..3c4dcc8b1 100644
--- a/docs/archive/specs/2026-04-02-improvement-hub-design.md
+++ b/docs/archive/specs/2026-04-02-improvement-hub-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Improvement Hub Redesign
-audience: [developer, analyst]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [improvement, prioritization, what-if, pdca, actions, verification, matrix, risk]
 ---
 
diff --git a/docs/archive/specs/2026-04-02-improvement-workspace-pdca-design.md b/docs/archive/specs/2026-04-02-improvement-workspace-pdca-design.md
index e2f43ff05..ca1f55ea6 100644
--- a/docs/archive/specs/2026-04-02-improvement-workspace-pdca-design.md
+++ b/docs/archive/specs/2026-04-02-improvement-workspace-pdca-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Improvement Workspace PDCA Extension
-audience: [developer, analyst]
+audience: human
 category: architecture
 status: superseded
 superseded-by: 2026-04-02-improvement-hub-design.md
diff --git a/docs/archive/specs/2026-04-02-unified-header-design.md b/docs/archive/specs/2026-04-02-unified-header-design.md
index 0f62803dd..ddfa6d964 100644
--- a/docs/archive/specs/2026-04-02-unified-header-design.md
+++ b/docs/archive/specs/2026-04-02-unified-header-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Unified Header — App + Project Merge
 date: 2026-04-02
-status: delivered
-audience: [developer, designer]
+status: active
+audience: human
 category: design-spec
 related: [header, navigation, auto-save, settings, responsive, workspace-tabs]
 supersedes: 2026-04-01-header-redesign-design.md (extends Phase 2)
diff --git a/docs/archive/specs/2026-04-02-web-first-deployment-architecture-design.md b/docs/archive/specs/2026-04-02-web-first-deployment-architecture-design.md
index 91ee5cc6c..65a98f158 100644
--- a/docs/archive/specs/2026-04-02-web-first-deployment-architecture-design.md
+++ b/docs/archive/specs/2026-04-02-web-first-deployment-architecture-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Web-First Deployment Architecture
-audience: [developer, architect, business]
+audience: human
 category: architecture
 status: draft
 related: [deployment, trust, enterprise, marketplace, collaboration, teams, storage, permissions]
diff --git a/docs/archive/specs/2026-04-02-web-first-implementation-design.md b/docs/archive/specs/2026-04-02-web-first-implementation-design.md
index cfc7eb0ad..afb7bbb92 100644
--- a/docs/archive/specs/2026-04-02-web-first-implementation-design.md
+++ b/docs/archive/specs/2026-04-02-web-first-implementation-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Web-First Implementation Design
-audience: [developer, architect]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [adr-059, adr-060, adr-026, adr-049, adr-033, web-first, knowledge-base, blob-storage]
 ---
 
diff --git a/docs/archive/specs/2026-04-03-hmw-brainstorm-modal-design.md b/docs/archive/specs/2026-04-03-hmw-brainstorm-modal-design.md
index b8c6769e4..8bb5fa50a 100644
--- a/docs/archive/specs/2026-04-03-hmw-brainstorm-modal-design.md
+++ b/docs/archive/specs/2026-04-03-hmw-brainstorm-modal-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: HMW Brainstorm Modal + Collaborative Sessions
-audience: [analyst, engineer]
+audience: human
 category: workflow
-status: delivered
+status: active
 related: [improvement, brainstorm, hmw, coscout, collaboration, ideation]
 ---
 
diff --git a/docs/archive/specs/2026-04-05-continuous-regression-design.md b/docs/archive/specs/2026-04-05-continuous-regression-design.md
index 23ed2aa93..38177b4f8 100644
--- a/docs/archive/specs/2026-04-05-continuous-regression-design.md
+++ b/docs/archive/specs/2026-04-05-continuous-regression-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Continuous Regression Design
-status: delivered
+status: active
 date: 2026-04-05
-audience: [engineer]
+audience: human
 category: analysis
 related: [best-subsets, factor-intelligence, evidence-map, what-if, ols, glm]
 ---
diff --git a/docs/archive/specs/2026-04-05-coscout-cognitive-redesign-design.md b/docs/archive/specs/2026-04-05-coscout-cognitive-redesign-design.md
index 68cd1aad7..959dbc06d 100644
--- a/docs/archive/specs/2026-04-05-coscout-cognitive-redesign-design.md
+++ b/docs/archive/specs/2026-04-05-coscout-cognitive-redesign-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: CoScout Cognitive Redesign
-status: delivered
+status: active
 date: 2026-04-05
 related:
   [
diff --git a/docs/archive/specs/2026-04-07-interaction-effects-design.md b/docs/archive/specs/2026-04-07-interaction-effects-design.md
index b79c131ee..eb90c4a81 100644
--- a/docs/archive/specs/2026-04-07-interaction-effects-design.md
+++ b/docs/archive/specs/2026-04-07-interaction-effects-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Interaction Effects — Two-Pass Best Subsets with OLS Interaction Screening
-status: delivered
+status: active
 date: 2026-04-07
 related:
   [
diff --git a/docs/archive/specs/2026-04-07-unified-whatif-explorer-design.md b/docs/archive/specs/2026-04-07-unified-whatif-explorer-design.md
index d4c2848e4..0391ce573 100644
--- a/docs/archive/specs/2026-04-07-unified-whatif-explorer-design.md
+++ b/docs/archive/specs/2026-04-07-unified-whatif-explorer-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Unified What-If Explorer
-status: delivered
+status: active
 date: 2026-04-07
 related:
   [
diff --git a/docs/archive/specs/2026-04-25-engagement-profile-raci-design.md b/docs/archive/specs/2026-04-25-engagement-profile-raci-design.md
index 0c465634d..62f21538f 100644
--- a/docs/archive/specs/2026-04-25-engagement-profile-raci-design.md
+++ b/docs/archive/specs/2026-04-25-engagement-profile-raci-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: 'Engagement-profile RACI — Identity and Engagement Layers (Brainstorm)'
-audience: [engineer, designer]
+audience: human
 category: architecture
 status: draft
 related:
diff --git a/docs/archive/specs/2026-04-25-process-hub-design.md b/docs/archive/specs/2026-04-25-process-hub-design.md
index 7c026b281..70d525345 100644
--- a/docs/archive/specs/2026-04-25-process-hub-design.md
+++ b/docs/archive/specs/2026-04-25-process-hub-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Process Hub - Work-System Context for Team Improvement
-audience: [product, designer, engineer]
+audience: human
 category: design-spec
-status: in-progress
+status: active
 related:
   [
     process-hub,
diff --git a/docs/archive/specs/2026-04-25-process-hub-use-cases.md b/docs/archive/specs/2026-04-25-process-hub-use-cases.md
index 48cd005b2..629a82cec 100644
--- a/docs/archive/specs/2026-04-25-process-hub-use-cases.md
+++ b/docs/archive/specs/2026-04-25-process-hub-use-cases.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Process Hub Use Cases - Process-First Operating Context
-audience: [product, designer, engineer]
+audience: human
 category: design-spec
-status: in-progress
+status: active
 related: [process-hub, evidence-sources, data-profiles, azure, portfolio, user-journeys]
 date: 2026-04-25
 ---
diff --git a/docs/archive/specs/2026-04-26-phase-6-sustainment-control-handoff-design.md b/docs/archive/specs/2026-04-26-phase-6-sustainment-control-handoff-design.md
index fd99d99e9..136ac1c0d 100644
--- a/docs/archive/specs/2026-04-26-phase-6-sustainment-control-handoff-design.md
+++ b/docs/archive/specs/2026-04-26-phase-6-sustainment-control-handoff-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: 'Phase 6 - Sustainment & Control Handoff'
-audience: [product, designer, engineer]
+audience: human
 category: design-spec
-status: delivered
+status: active
 related:
   - 2026-04-26-unified-process-hub-methodology-roadmap.md
   - 2026-04-25-process-hub-design.md
diff --git a/docs/archive/specs/2026-04-26-unified-process-hub-methodology-roadmap.md b/docs/archive/specs/2026-04-26-unified-process-hub-methodology-roadmap.md
index f88f5a2f2..0e803cf52 100644
--- a/docs/archive/specs/2026-04-26-unified-process-hub-methodology-roadmap.md
+++ b/docs/archive/specs/2026-04-26-unified-process-hub-methodology-roadmap.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: Unified Process Hub Methodology Roadmap
-audience: [product, designer, engineer]
+audience: human
 category: design-spec
-status: in-progress
+status: active
 related:
   [
     process-hub,
diff --git a/docs/archive/specs/2026-04-27-actionable-current-process-state-panel-design.md b/docs/archive/specs/2026-04-27-actionable-current-process-state-panel-design.md
index c03a8193c..e5d8cd11e 100644
--- a/docs/archive/specs/2026-04-27-actionable-current-process-state-panel-design.md
+++ b/docs/archive/specs/2026-04-27-actionable-current-process-state-panel-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Actionable Current Process State Panel
-audience: engineer
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/archive/specs/2026-04-27-layered-process-view-design.md b/docs/archive/specs/2026-04-27-layered-process-view-design.md
index a032beae1..9b472a78d 100644
--- a/docs/archive/specs/2026-04-27-layered-process-view-design.md
+++ b/docs/archive/specs/2026-04-27-layered-process-view-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Layered Process View — Three-Band Visual Language
-audience: [product, designer, engineer, analyst]
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/archive/specs/2026-04-27-process-learning-operating-model-design.md b/docs/archive/specs/2026-04-27-process-learning-operating-model-design.md
index 85b22c484..1272cc1db 100644
--- a/docs/archive/specs/2026-04-27-process-learning-operating-model-design.md
+++ b/docs/archive/specs/2026-04-27-process-learning-operating-model-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Process Learning Operating Model (SUPERSEDED)
-audience: [product, designer, engineer, analyst]
+audience: human
 category: design-spec
 status: superseded
 superseded-by: docs/superpowers/specs/2026-05-03-variscout-vision-design.md
diff --git a/docs/archive/specs/2026-04-27-product-method-roadmap-design.md b/docs/archive/specs/2026-04-27-product-method-roadmap-design.md
index 69c44962b..22a64352c 100644
--- a/docs/archive/specs/2026-04-27-product-method-roadmap-design.md
+++ b/docs/archive/specs/2026-04-27-product-method-roadmap-design.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Product-Method Roadmap (SUPERSEDED)
-audience: [product, designer, engineer, analyst, manager]
+audience: human
 category: design-spec
 status: superseded
 superseded-by: docs/superpowers/specs/2026-05-03-variscout-vision-design.md
diff --git a/docs/archive/specs/2026-04-29-customer-tenant-ingestion-rollups-concept.md b/docs/archive/specs/2026-04-29-customer-tenant-ingestion-rollups-concept.md
index c6baad493..fba48518c 100644
--- a/docs/archive/specs/2026-04-29-customer-tenant-ingestion-rollups-concept.md
+++ b/docs/archive/specs/2026-04-29-customer-tenant-ingestion-rollups-concept.md
@@ -1,6 +1,8 @@
 ---
+tier: stable
+purpose: remember
 title: Customer-Tenant Ingestion And Rollups Concept
-audience: [product, architect, engineer, admin]
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/archive/specs/2026-05-03-framing-layer-design.md b/docs/archive/specs/2026-05-03-framing-layer-design.md
index 4fe8227c9..0f1809bf1 100644
--- a/docs/archive/specs/2026-05-03-framing-layer-design.md
+++ b/docs/archive/specs/2026-05-03-framing-layer-design.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: remember
 title: VariScout Framing Layer — Hub Creation, Data Ingestion, and Investigation Entry
-audience: [product, engineer, designer]
+audience: human
 category: design-spec
-status: delivered
+status: active
 last-reviewed: 2026-05-04
 related:
   - docs/superpowers/specs/2026-05-03-variscout-vision-design.md
diff --git a/docs/decision-log.md b/docs/decision-log.md
index dfe50737c..28a929c77 100644
--- a/docs/decision-log.md
+++ b/docs/decision-log.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: decide
 title: VariScout — Decision Log
-audience: [product, designer, engineer, analyst]
+audience: human
 category: living-index
-status: living
+status: active
 last-reviewed: 2026-05-07
 ---
 
@@ -26,7 +28,7 @@ When in doubt: capture, don't invent. Record the decision; link to its source ar
 
 Decisions we keep relitigating. Each entry: short statement, rationale, closing artifact, date pinned.
 
-- **2026-05-16 — Wedge pivot: single-product VariScout for improvement specialists (€99/mo, 6-tab nav, project-membership ACLs).** After an 8-round strategic brainstorm questioning whether VariScout was trying to be too many things at once (Hub + Project containers, 4 personas, tier-gating across 28 files, multi-source pipeline plans, process-owner monitoring concepts), the conversation converged on splitting VariScout into **two products on a roadmap**: ship the wedge first (project tool for improvement specialists), defer the platform (process ownership, auto pipelines, Hub portfolios, 4 personas) to a future separate product called **VariScout Process**. V1 (this pivot) anatomy: (1) **6 tabs in workflow order** — `[Home] [Projects] [Process] [Analyze] [Investigation] [Report]` — Improve tab removed as top-level; becomes a stage inside Projects detail; supersedes the Coherence §4 / Projects-tab-design D1 7-tab nav and Coherence's verb/noun grouping rationale (workflow-order grouping is stronger for the wedge's single persona). (2) **Projects detail = 4 stages** — `Charter / Approach / Improve / Sustainment` (Handoff folds into Sustainment closure as "did it work? + close"); supersedes Projects-tab-design D-series stage list (Sustainment+Handoff → Improve+Sustainment). (3) **Improve stage UI = simple action tracker by default**; PDCA workbench + What-If accessible via an "Advanced" toggle (progressive disclosure, not parallel modes). What-If may re-emerge in Analyze later; idea board / action conversion retire. (4) **3 canvas response paths** at V1 (Investigate, Quick Action, Charter); Sustainment auto-fires per ADR-080; Handoff path deleted everywhere. (5) **Persona model collapses to one — Specialist** — `personaRole` routing deleted from V1; Coherence Session A's 4-persona model migrates to VariScout Process. Member-roles inside projects (Lead / Member / Reviewer) replace persona routing. (6) **Per-project membership ACLs** — Project Lead invites org users _per project_ (not tenant-wide); cross-Azure-AD-tenant invites explicitly out of V1 scope (privacy boundary as feature, not bug); ACL data model: `ProjectMember`, `Invitation`, role enum. (7) **Pricing: single €99/month SKU, Azure tenant-wide, unlimited org users + projects**; supersedes ADR-033's €79/€199 tier split; preserves ADR-033 H6 hypothesis ("per-deployment beats per-seat") — €99 is still per-deployment. Tier-gating logic (~28 files using `isPaidTier()` / `hasTeamFeatures()`) mostly retires; project-membership ACL checks replace where access-gating remains useful. (8) **VariScout Process** named-future for process-ownership platform features (Hub portfolios, 4 personas, auto pipelines, multi-source ingestion). Not announced in V1 marketing; mentioned only when customers ask about enterprise use cases. (9) **Positioning unchanged**: "Structured investigation for process improvement" — V1 delivers the whole sentence for one project lead and their invited team. (10) **Three preconditions before engineering commits**: migration math (30-min financial sensitivity on €79→€99 + €199→€99), Azure AD invitation constraint accepted as feature (confirms cross-org collab is out of V1 ICP), one customer validation conversation (30-min pitch confirms "yes that's the product I want"). (11) **Holistic doc cleanup (~500 docs)** sequenced as Audit → Triage → Apply (Phase C), with canonical anchor docs (OVERVIEW, USER-JOURNEYS, DATA-FLOW, positioning, business-bible, tier-philosophy, roadmap, decision-log, root CLAUDE.md, llms.txt) updated as Phase A before engineering; per-surface amendments as Phase B during engineering; archive over delete to preserve institutional knowledge per `feedback_consolidation_replace_not_umbrella` + `feedback_close_threads_to_done`. (12) **Recent week's shipped work (PRs #172–#181) is mostly wedge-aligned** — already project-centric. Persona-routing hooks need de-personalization (small refactor, not rewrite); stage list rename (Sustainment+Handoff → Improve+Sustainment) is a medium refactor; Improve top-level tab deletion is the largest engineering item. Canonical artifacts: spec [`docs/superpowers/specs/2026-05-16-wedge-architecture-design.md`](superpowers/specs/2026-05-16-wedge-architecture-design.md) (canonical V1 design); ADR [`docs/07-decisions/adr-082-wedge-architecture.md`](07-decisions/adr-082-wedge-architecture.md). Supersession markers added to ADR-007, ADR-033, Coherence design spec, Projects-tab design spec. Brainstorm captured in plan file `~/.claude/plans/i-am-wondering-that-smooth-narwhal.md`. _Pinned 2026-05-16._
+- **2026-05-16 — Wedge pivot: single-product VariScout for improvement specialists (€99/mo, 6-tab nav, project-membership ACLs).** VariScout splits into two products on a roadmap — ship the wedge (project tool for improvement specialists) first; defer the platform (process ownership, Hub portfolios, 4-persona model, auto pipelines) to **VariScout Process**, a separately-scoped future product. V1 collapses to one persona (Specialist) with project-membership roles (Lead / Member / Sponsor); 6-tab workflow nav `[Home] [Projects] [Process] [Analyze] [Investigation] [Report]`; Improve becomes a stage inside Projects detail (Handoff folds into Sustainment); 3 canvas response paths (Investigate, Quick Action, Charter); €99/month single Azure SKU honors ADR-033's H6 per-deployment hypothesis; cross-AD-tenant invites explicitly out of V1 scope. Three preconditions block engineering commit: migration math (€79→€99 + €199→€99 financial sensitivity), Azure AD invitation constraint acceptance, one customer validation conversation. Holistic doc cleanup of ~500 docs evolves from a Phase C cliff into a continuous Steward loop per the docs-strategy-2026 plan; archive over delete preserves institutional knowledge. **Canonical:** [ADR-082](07-decisions/adr-082-wedge-architecture.md) (decision record — see §Decision for the V1 surface table and §Pricing rationale) + [wedge spec](superpowers/specs/2026-05-16-wedge-architecture-design.md) (V1 design). Supersession markers on ADR-007, ADR-033, Coherence spec, Projects-tab spec. Brainstorm transcript: `~/.claude/plans/i-am-wondering-that-smooth-narwhal.md`. _Pinned 2026-05-16._
 
 - **2026-05-14 — Projects tab + IP detail (Lifecycle) page + Home active-IP launchpad locked.** Brainstorm session resolved coherence spec §15's "IP detail page details" carry-forward and reframed the top-nav structure. Six locks: (1) **7-tab nav split** — coherence's locked 6-tab Path A becomes 7 tabs (`[Home] [Process] [Analyze] [Investigation] [Improve] [Projects] [Report]`), splitting the previous single "Improvement" tab into a verb tab (Improve = legacy `ImprovementView` / PDCA workbench, zero migration) and a noun tab (Projects = new IP lifecycle + detail page); deliberate verb/noun split because the two surfaces serve genuinely different jobs (casual ideation across IPs vs. tracked-project authoring). (2) **Home becomes the active-IP launchpad** for the Project Lead persona; picking an active IP at Home cascades scope to Process / Analyze / Investigation / Improve / Report tabs via the IP-context chip (Coherence §11 pattern #5). Process Owner doesn't pick (hub-level cadence work); SME auto-scopes when accepting a consult; Frontline lives in action cards. State persists per-Hub-per-user via `localStorage` (new `useActiveIPStore` Annotation-layer Zustand store). (3) **IP detail (Lifecycle) page anatomy** — header (back-link / status pill / title / goal summary / team avatars + Invite) + stage tabs (`Charter / Approach / Sustainment / Handoff`) + Overview/Sections segmented toggle (Coherence §11 pattern #2) + 280px right rail team workspace. (4) **Approach stage is SuspectedCause-anchored** — per-cause hierarchy (Hypothesis → ImprovementIdea → ActionItem) as the visual unit in both Overview and Sections modes. Sections mode shows read-only summary with per-cause "Open in Improve workbench" jump-out (avoids duplicating the legacy PDCA tooling). (5) **Report tab IP-scoped with Overview/Technical audience toggle** (fractal pattern with IP detail's mode toggle) — Overview renders a 7-section QC-Story-shaped narrative arc with plain-English UI copy ("Where we started" / "What we aimed for" / "What we found + what we did" / "Did it work?" / "What we standardized + learned" / "What's next"); methodology lineage stays internal per RPS V1 D2 (`feedback_drop_methodology_bridges`). Technical layer renders the full analytical chart suite. Free-roaming Report = Hub-level portfolio. (6) **Single data-model addition** — optional `ImprovementProject.reflection?: string` field (sibling of `signoff`) for analyst lessons-learned narrative; backward-compatible, no `.vrs` format version bump. Spec [`docs/superpowers/specs/2026-05-14-projects-tab-design.md`](superpowers/specs/2026-05-14-projects-tab-design.md) written + committed as `2f4cff53`. Coherence spec amendments applied: §4 (nav: 6 → 7 tabs + Improve/Projects rows in table + amendment block), §15 (carry-forward marked resolved), §16 (new row in spec contributions matrix). Implementation plans next (4 sub-plans recommended: Projects tab + IP detail / Home + cascade / Team rail / Report). Tier behavior: V1 features in PWA free + Azure paid; V2 collaboration features (threaded comments, @-mentions, signoff queue UI, RACI per-section, change notifications) stay deferred + tier-gated when shipped per `feedback_tier_gate_inside_surface`. _Pinned 2026-05-14._
 
diff --git a/docs/glossary.md b/docs/glossary.md
index a0e5980e9..b880d4e4e 100644
--- a/docs/glossary.md
+++ b/docs/glossary.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: 'Glossary'
-audience: [analyst, engineer]
+audience: human
 category: learning
-status: stable
+status: active
 ---
 
 # Glossary
diff --git a/docs/index.md b/docs/index.md
index e76a48e61..3135cc93a 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -1,9 +1,11 @@
 ---
+tier: stable
+purpose: orient
 title: VariScout Documentation
 description: Structured investigation for process improvement — question-driven, evidence-based, AI-assisted
-audience: [developer, architect]
+audience: human
 category: reference
-status: stable
+status: active
 ---
 
 # VariScout Documentation
diff --git a/docs/investigations.md b/docs/investigations.md
index b35712465..b81677bf7 100644
--- a/docs/investigations.md
+++ b/docs/investigations.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: decide
 title: VariScout — Things to Investigate
-audience: [engineer, designer, analyst]
+audience: human
 category: living-index
-status: living
+status: active
 last-reviewed: 2026-05-03
 ---
 
@@ -622,3 +624,33 @@ Until then: stays as a logged investigation. The current tripwire remains the en
 **Resolution [2026-05-08]:** `computeHistogramBins(values, rule?)` helper added in `@variscout/core/stats` (Sturges default, Scott option). `CanvasStepMiniChart` histogram branch now renders one bar per bin with bin counts as heights, replacing the first-12-raw-values normalization. Empty bins floor at 8% height so the bin axis stays legible — this replaces the prior 15% floor (which was tuned for 12 raw-value pseudo-bars; 8% reads more honestly when bin counts can legitimately be zero).
 
 ---
+
+### Enforcement gap: `Math.random()` ban has no ESLint rule
+
+**Surfaced by:** Play 6 invariant consolidation, 2026-05-16.
+
+**Description:** The `Math.random()` ban is listed as a hard invariant in `.claude/INVARIANTS.md` and enforced by convention only. ESLint rules `no-root-cause-language` and `no-interaction-moderator` exist as precedents. A scoped `no-math-random` rule would close the gap mechanically.
+
+**Possible directions:**
+
+- Add `variscout/no-math-random` to `tools/eslint-plugin-variscout/` scoped to `packages/core/src/**` + `**/__tests__/**` + `packages/*/src/**`. Exclude `e2e/` where pseudo-randomness may be intentional.
+- Alt: `no-restricted-globals` with `Math.random` message pointing to the seeded PRNG helper.
+
+**Promotion path:** small PR — plugin rule + ESLint config wiring. Precedent is the existing plugin rules in `tools/eslint-plugin-variscout/`.
+
+---
+
+### Enforcement gap: Tailwind `@source` directive coverage has no automated check
+
+**Surfaced by:** Play 6 invariant consolidation, 2026-05-16.
+
+**Description:** The invariant requiring `@source` directives in every `apps/*/src/index.css` for shared packages is enforced by convention + visual regression only. Missing directives silently break responsive utilities (Tailwind v4 behaviour). A build-time check would surface the gap deterministically.
+
+**Possible directions:**
+
+- As part of Play 7 (`pnpm docs:gen-arch`): extend the script to emit a Tailwind source map check — enumerate `packages/*/` and verify each appears in all `apps/*/src/index.css` `@source` lists.
+- Simpler: a `scripts/check-tailwind-sources.sh` that `grep`s each package name against the app CSS files and fails if any are missing.
+
+**Promotion path:** Play 7 candidate (system auto-generation). Or standalone script in a cleanup PR.
+
+---
diff --git a/docs/roadmap.md b/docs/roadmap.md
index f755d8a28..7bc2b45a4 100644
--- a/docs/roadmap.md
+++ b/docs/roadmap.md
@@ -1,8 +1,10 @@
 ---
+tier: stable
+purpose: orient
 title: VariScout — Roadmap (lightweight, post-canvas-migration)
-audience: [product, engineer, designer]
+audience: human
 category: living-index
-status: living
+status: active
 last-reviewed: 2026-05-16
 related:
   - docs/decision-log.md
diff --git a/docs/superpowers/plans/2026-04-02-coscout-intelligence-architecture.md b/docs/superpowers/plans/2026-04-02-coscout-intelligence-architecture.md
index 851034594..43bebddf0 100644
--- a/docs/superpowers/plans/2026-04-02-coscout-intelligence-architecture.md
+++ b/docs/superpowers/plans/2026-04-02-coscout-intelligence-architecture.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: CoScout Intelligence Architecture (ADR-060) Implementation Plan
-status: in-progress
+status: active
 spec: 2026-04-02-coscout-intelligence-architecture-design.md
 ---
 
diff --git a/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase1-foundation.md b/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase1-foundation.md
index 077e6370b..41e2bf0c9 100644
--- a/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase1-foundation.md
+++ b/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase1-foundation.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'Agent Docs Architecture — Phase 1: Foundation'
-status: delivered
+status: active
 date: 2026-04-17
 ---
 
diff --git a/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase2-migration.md b/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase2-migration.md
index aa6ee0184..cec644f5b 100644
--- a/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase2-migration.md
+++ b/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase2-migration.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'Agent Docs Architecture — Phase 2: Migration'
-status: delivered
+status: active
 date: 2026-04-17
 ---
 
diff --git a/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase3-enforcement.md b/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase3-enforcement.md
index 538aebee5..64a08cd99 100644
--- a/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase3-enforcement.md
+++ b/docs/superpowers/plans/2026-04-17-agent-docs-architecture-phase3-enforcement.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'Agent Docs Architecture — Phase 3: Enforcement'
-status: delivered
+status: active
 date: 2026-04-17
 ---
 
diff --git a/docs/superpowers/plans/2026-04-19-investigation-wall.md b/docs/superpowers/plans/2026-04-19-investigation-wall.md
index bd6a9a3c8..93329d18a 100644
--- a/docs/superpowers/plans/2026-04-19-investigation-wall.md
+++ b/docs/superpowers/plans/2026-04-19-investigation-wall.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Investigation Wall Implementation Plan
-status: in-progress
+status: active
 date: 2026-04-19
 ---
 
diff --git a/docs/superpowers/plans/2026-04-25-qde-2-branch-workspace.md b/docs/superpowers/plans/2026-04-25-qde-2-branch-workspace.md
index 8f12c8a09..f24b516d5 100644
--- a/docs/superpowers/plans/2026-04-25-qde-2-branch-workspace.md
+++ b/docs/superpowers/plans/2026-04-25-qde-2-branch-workspace.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: QDE 2.0 Branch Workspace Phase Plan
-audience: [engineer, product]
+audience: human
 category: implementation
-status: delivered
+status: active
 date: 2026-04-25
 related: [question-driven-eda-2, mechanism-branches, investigation-wall]
 ---
diff --git a/docs/superpowers/plans/2026-04-25-question-driven-eda-2-handoff.md b/docs/superpowers/plans/2026-04-25-question-driven-eda-2-handoff.md
index e4a8b5229..97bba690c 100644
--- a/docs/superpowers/plans/2026-04-25-question-driven-eda-2-handoff.md
+++ b/docs/superpowers/plans/2026-04-25-question-driven-eda-2-handoff.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Question-Driven EDA 2.0 — Session Handoff
-audience: [engineer, product]
+audience: human
 category: implementation
-status: in-progress
+status: active
 date: 2026-04-25
 related:
   [
diff --git a/docs/superpowers/plans/2026-04-26-qde-2-survey-foundation.md b/docs/superpowers/plans/2026-04-26-qde-2-survey-foundation.md
index d71147a73..3201ed737 100644
--- a/docs/superpowers/plans/2026-04-26-qde-2-survey-foundation.md
+++ b/docs/superpowers/plans/2026-04-26-qde-2-survey-foundation.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: QDE 2.0 Survey Foundation Phase Plan
-audience: [engineer, product]
+audience: human
 category: implementation
-status: delivered
+status: active
 date: 2026-04-26
 related: [question-driven-eda-2, survey, mechanism-branches, signal-cards, process-map]
 ---
diff --git a/docs/superpowers/plans/2026-04-27-actionable-current-process-state-panel-plan.md b/docs/superpowers/plans/2026-04-27-actionable-current-process-state-panel-plan.md
index d7377f203..1fa10499a 100644
--- a/docs/superpowers/plans/2026-04-27-actionable-current-process-state-panel-plan.md
+++ b/docs/superpowers/plans/2026-04-27-actionable-current-process-state-panel-plan.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Actionable Current Process State Panel — Implementation Plan
-status: delivered
+status: active
 ---
 
 # Actionable Current Process State Panel — Implementation Plan
diff --git a/docs/superpowers/plans/2026-04-27-layered-process-view-v1.md b/docs/superpowers/plans/2026-04-27-layered-process-view-v1.md
index c6e165079..e5b1a2e45 100644
--- a/docs/superpowers/plans/2026-04-27-layered-process-view-v1.md
+++ b/docs/superpowers/plans/2026-04-27-layered-process-view-v1.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Layered Process View V1 Implementation Plan
-status: delivered
+status: active
 ---
 
 # Layered Process View V1 Implementation Plan
diff --git a/docs/superpowers/plans/2026-04-27-phase-3-h1-closure-h2-launch-plan.md b/docs/superpowers/plans/2026-04-27-phase-3-h1-closure-h2-launch-plan.md
index dd48ae102..ed9d8e45a 100644
--- a/docs/superpowers/plans/2026-04-27-phase-3-h1-closure-h2-launch-plan.md
+++ b/docs/superpowers/plans/2026-04-27-phase-3-h1-closure-h2-launch-plan.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Phase 3 — H1 Closure + H2 Launch — Implementation Plan
-status: delivered
+status: active
 ---
 
 # Phase 3 — H1 Closure + H2 Launch — Implementation Plan
diff --git a/docs/superpowers/plans/2026-04-28-production-line-glance-c1-data-and-hub-tab.md b/docs/superpowers/plans/2026-04-28-production-line-glance-c1-data-and-hub-tab.md
index 867f610a2..29b6f4d8a 100644
--- a/docs/superpowers/plans/2026-04-28-production-line-glance-c1-data-and-hub-tab.md
+++ b/docs/superpowers/plans/2026-04-28-production-line-glance-c1-data-and-hub-tab.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Production-Line-Glance — C1 Data Layer + Hub Capability Tab Implementation Plan
-audience: [engineer, architect]
+audience: human
 category: implementation
-status: delivered
+status: active
 related:
   [
     production-line-glance-surface-wiring-design,
diff --git a/docs/superpowers/plans/2026-04-28-production-line-glance-c2-layered-view.md b/docs/superpowers/plans/2026-04-28-production-line-glance-c2-layered-view.md
index 0f11dcbe5..47317e70d 100644
--- a/docs/superpowers/plans/2026-04-28-production-line-glance-c2-layered-view.md
+++ b/docs/superpowers/plans/2026-04-28-production-line-glance-c2-layered-view.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Production-Line-Glance — C2 LayeredProcessView Operations Band Implementation Plan
-audience: [engineer, architect]
+audience: human
 category: implementation
-status: delivered
+status: active
 related:
   [
     production-line-glance-surface-wiring-design,
diff --git a/docs/superpowers/plans/2026-04-28-production-line-glance-charts.md b/docs/superpowers/plans/2026-04-28-production-line-glance-charts.md
index dd35912d8..81f963b29 100644
--- a/docs/superpowers/plans/2026-04-28-production-line-glance-charts.md
+++ b/docs/superpowers/plans/2026-04-28-production-line-glance-charts.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Production-Line-Glance — Chart Components Layer (Plan B)
-audience: [engineer, architect]
+audience: human
 category: implementation
-status: delivered
+status: active
 related: [production-line-glance-design, production-line-glance-engine]
 date: 2026-04-28
 ---
diff --git a/docs/superpowers/plans/2026-04-28-production-line-glance-engine.md b/docs/superpowers/plans/2026-04-28-production-line-glance-engine.md
index 92e72d8b6..4ff3098ab 100644
--- a/docs/superpowers/plans/2026-04-28-production-line-glance-engine.md
+++ b/docs/superpowers/plans/2026-04-28-production-line-glance-engine.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Production-Line-Glance — Engine Layer (Plan A)
-status: delivered
+status: active
 ---
 
 # Production-Line-Glance — Engine Layer (Plan A)
diff --git a/docs/superpowers/plans/2026-04-29-multi-level-scout-v1-decisions.md b/docs/superpowers/plans/2026-04-29-multi-level-scout-v1-decisions.md
index 288c013ec..5dfffca26 100644
--- a/docs/superpowers/plans/2026-04-29-multi-level-scout-v1-decisions.md
+++ b/docs/superpowers/plans/2026-04-29-multi-level-scout-v1-decisions.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'Multi-level SCOUT V1 — load-bearing decisions'
-audience: [engineer, architect]
+audience: human
 category: implementation
-status: accepted
+status: active
 date: 2026-04-29
 related:
   - multi-level-scout-design
diff --git a/docs/superpowers/plans/2026-04-29-multi-level-scout-v1.md b/docs/superpowers/plans/2026-04-29-multi-level-scout-v1.md
index 64ecb078e..0ec15b667 100644
--- a/docs/superpowers/plans/2026-04-29-multi-level-scout-v1.md
+++ b/docs/superpowers/plans/2026-04-29-multi-level-scout-v1.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'Multi-level SCOUT V1 (first slice) — implementation plan'
-audience: [engineer, architect]
+audience: human
 category: implementation
-status: in-progress
+status: active
 date: 2026-04-29
 related:
   - multi-level-scout-design
diff --git a/docs/superpowers/plans/2026-05-03-framing-layer-v1-slice-1-decisions.md b/docs/superpowers/plans/2026-05-03-framing-layer-v1-slice-1-decisions.md
index 916411b50..12c15743b 100644
--- a/docs/superpowers/plans/2026-05-03-framing-layer-v1-slice-1-decisions.md
+++ b/docs/superpowers/plans/2026-05-03-framing-layer-v1-slice-1-decisions.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Framing Layer V1 Slice 1 — Locked Decisions
-audience: [engineer]
+audience: human
 category: implementation
 status: draft
 date: 2026-05-03
diff --git a/docs/superpowers/plans/2026-05-03-framing-layer-v1-slice-1.md b/docs/superpowers/plans/2026-05-03-framing-layer-v1-slice-1.md
index 4ee7136a9..e3b309340 100644
--- a/docs/superpowers/plans/2026-05-03-framing-layer-v1-slice-1.md
+++ b/docs/superpowers/plans/2026-05-03-framing-layer-v1-slice-1.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Framing Layer V1 Slice 1 — Foundation (Mode B + Mode A.1 + PWA opt-in persistence)
-audience: [engineer]
+audience: human
 category: implementation
 status: draft
 date: 2026-05-03
diff --git a/docs/superpowers/plans/2026-05-03-scout-ui-consolidation.md b/docs/superpowers/plans/2026-05-03-scout-ui-consolidation.md
index 671cdf4fe..fcd79b252 100644
--- a/docs/superpowers/plans/2026-05-03-scout-ui-consolidation.md
+++ b/docs/superpowers/plans/2026-05-03-scout-ui-consolidation.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Scout UI / Navigation consolidation
-status: delivered
+status: active
 ---
 
 # Scout UI / Navigation consolidation
diff --git a/docs/superpowers/plans/2026-05-04-canvas-manual-authoring-pr4.md b/docs/superpowers/plans/2026-05-04-canvas-manual-authoring-pr4.md
index 9d38a15e1..16682fc1c 100644
--- a/docs/superpowers/plans/2026-05-04-canvas-manual-authoring-pr4.md
+++ b/docs/superpowers/plans/2026-05-04-canvas-manual-authoring-pr4.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Manual Canvas Authoring (PR4) Implementation Plan
-audience: [engineer]
+audience: human
 category: implementation
 status: draft
 date: 2026-05-04
diff --git a/docs/superpowers/plans/2026-05-06-canvas-pr4c-pr6-followup.md b/docs/superpowers/plans/2026-05-06-canvas-pr4c-pr6-followup.md
index b1fc149c9..2dc85f1c3 100644
--- a/docs/superpowers/plans/2026-05-06-canvas-pr4c-pr6-followup.md
+++ b/docs/superpowers/plans/2026-05-06-canvas-pr4c-pr6-followup.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Canvas PR4c-PR6 Retrospective Followup — fix list + vision-alignment plan
-audience: [engineer]
+audience: human
 category: implementation
 status: draft
 date: 2026-05-06
diff --git a/docs/superpowers/plans/2026-05-06-data-flow-foundation-f1-f2-audit.md b/docs/superpowers/plans/2026-05-06-data-flow-foundation-f1-f2-audit.md
index 239e23d9d..ebc509c9d 100644
--- a/docs/superpowers/plans/2026-05-06-data-flow-foundation-f1-f2-audit.md
+++ b/docs/superpowers/plans/2026-05-06-data-flow-foundation-f1-f2-audit.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: F1+F2 Audit — entity / factory / persistence / store / canvas-action inventories
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-06
diff --git a/docs/superpowers/plans/2026-05-06-data-flow-foundation-f1-f2.md b/docs/superpowers/plans/2026-05-06-data-flow-foundation-f1-f2.md
index 7fc65521a..bb8671946 100644
--- a/docs/superpowers/plans/2026-05-06-data-flow-foundation-f1-f2.md
+++ b/docs/superpowers/plans/2026-05-06-data-flow-foundation-f1-f2.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Data-Flow Foundation F1+F2 Implementation Plan — type-level normalization + repository pattern
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-06
diff --git a/docs/superpowers/plans/2026-05-06-data-flow-foundation-f3-5-ingestion-audit.md b/docs/superpowers/plans/2026-05-06-data-flow-foundation-f3-5-ingestion-audit.md
index 967ab2bef..56f9d7c2d 100644
--- a/docs/superpowers/plans/2026-05-06-data-flow-foundation-f3-5-ingestion-audit.md
+++ b/docs/superpowers/plans/2026-05-06-data-flow-foundation-f3-5-ingestion-audit.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Data-Flow Foundation F3.5 Ingestion Audit
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-06
diff --git a/docs/superpowers/plans/2026-05-06-data-flow-foundation-f3-5-ingestion.md b/docs/superpowers/plans/2026-05-06-data-flow-foundation-f3-5-ingestion.md
index 5d272728d..42bf5ddc6 100644
--- a/docs/superpowers/plans/2026-05-06-data-flow-foundation-f3-5-ingestion.md
+++ b/docs/superpowers/plans/2026-05-06-data-flow-foundation-f3-5-ingestion.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Data-Flow Foundation F3.5 — Ingestion Action Layer (paste / upload unified onto EVIDENCE_ADD_SNAPSHOT dispatch)
-audience: [engineer]
+audience: human
 category: implementation
-status: delivered
+status: active
 last-reviewed: 2026-05-07
 related:
   - docs/superpowers/specs/2026-05-06-data-flow-foundation-design.md
diff --git a/docs/superpowers/plans/2026-05-07-canvas-pr8-8a-mode-aware-ctas.md b/docs/superpowers/plans/2026-05-07-canvas-pr8-8a-mode-aware-ctas.md
index 7a11f2b9a..b4c906ccd 100644
--- a/docs/superpowers/plans/2026-05-07-canvas-pr8-8a-mode-aware-ctas.md
+++ b/docs/superpowers/plans/2026-05-07-canvas-pr8-8a-mode-aware-ctas.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Canvas PR8 sub-PR 8a — Mode-aware response-path CTAs
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/plans/2026-05-07-canvas-pr8-8b-drift-and-time-series.md b/docs/superpowers/plans/2026-05-07-canvas-pr8-8b-drift-and-time-series.md
index 4a367357a..c5090dcbd 100644
--- a/docs/superpowers/plans/2026-05-07-canvas-pr8-8b-drift-and-time-series.md
+++ b/docs/superpowers/plans/2026-05-07-canvas-pr8-8b-drift-and-time-series.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Canvas PR8 sub-PR 8b — Drift indicator + time-series mini-chart
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/plans/2026-05-07-canvas-pr8-8d-hypothesis-arrow-drawing.md b/docs/superpowers/plans/2026-05-07-canvas-pr8-8d-hypothesis-arrow-drawing.md
index d91df4bb1..46ad704c4 100644
--- a/docs/superpowers/plans/2026-05-07-canvas-pr8-8d-hypothesis-arrow-drawing.md
+++ b/docs/superpowers/plans/2026-05-07-canvas-pr8-8d-hypothesis-arrow-drawing.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'Canvas Hypothesis-Arrow Drawing — PR8 sub-PR 8d implementation plan'
-audience: [engineer, product]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/plans/2026-05-07-canvas-pr8-vision-alignment-master.md b/docs/superpowers/plans/2026-05-07-canvas-pr8-vision-alignment-master.md
index f4f5137d0..9ed3b523b 100644
--- a/docs/superpowers/plans/2026-05-07-canvas-pr8-vision-alignment-master.md
+++ b/docs/superpowers/plans/2026-05-07-canvas-pr8-vision-alignment-master.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Canvas Migration PR8 — Vision Alignment Master Plan (sequencing 8a / 8b / 8d / 8e + 8f deferral)
-audience: [engineer, product]
+audience: human
 category: implementation
-status: delivered
+status: active
 last-reviewed: 2026-05-08
 related:
   - docs/superpowers/specs/2026-05-04-canvas-migration-design.md
diff --git a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-azure-provenance-parity-audit.md b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-azure-provenance-parity-audit.md
index 818e575a6..e78727c94 100644
--- a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-azure-provenance-parity-audit.md
+++ b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-azure-provenance-parity-audit.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Data-Flow Foundation F3.6-β Audit — Azure persistence + cloud-sync surface
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-azure-provenance-parity.md b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-azure-provenance-parity.md
index c5951e49c..4d063582b 100644
--- a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-azure-provenance-parity.md
+++ b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-azure-provenance-parity.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Data-Flow Foundation F3.6-β — Azure rowProvenance parity (envelope facet + cloud-sync via Blob Storage + ETag concurrency)
-audience: [engineer]
+audience: human
 category: implementation
-status: delivered
+status: active
 last-reviewed: 2026-05-07
 related:
   - docs/superpowers/specs/2026-05-06-data-flow-foundation-design.md
diff --git a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-chrome-walk.md b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-chrome-walk.md
index baa771982..4c4c838c1 100644
--- a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-chrome-walk.md
+++ b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f3-6-chrome-walk.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: F3.6-β — Chrome walk script (Azure persistence + cloud sync + ETag concurrency)
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f4-audit.md b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f4-audit.md
index b87b251fa..55e2ddb7b 100644
--- a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f4-audit.md
+++ b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f4-audit.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: F4 audit — touch surface verified pre-implementation
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f4-three-layer-state.md b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f4-three-layer-state.md
index c07a60300..0ced9fff4 100644
--- a/docs/superpowers/plans/2026-05-07-data-flow-foundation-f4-three-layer-state.md
+++ b/docs/superpowers/plans/2026-05-07-data-flow-foundation-f4-three-layer-state.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: F4 implementation plan — three-layer state codification
-audience: [engineer]
+audience: human
 category: implementation
 status: draft
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/plans/2026-05-08-canvas-polish-v1.md b/docs/superpowers/plans/2026-05-08-canvas-polish-v1.md
index 64bef265b..b7980f8f2 100644
--- a/docs/superpowers/plans/2026-05-08-canvas-polish-v1.md
+++ b/docs/superpowers/plans/2026-05-08-canvas-polish-v1.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Canvas Polish v1 — bundled PR1–PR9 follow-ups
-audience: [engineer]
+audience: human
 category: implementation
-status: delivered
+status: active
 related:
   - docs/investigations.md
   - docs/superpowers/specs/2026-05-08-canvas-wall-overlay-design.md
diff --git a/docs/superpowers/plans/2026-05-08-canvas-pr8-8e-wall-overlay.md b/docs/superpowers/plans/2026-05-08-canvas-pr8-8e-wall-overlay.md
index f03114238..41c772975 100644
--- a/docs/superpowers/plans/2026-05-08-canvas-pr8-8e-wall-overlay.md
+++ b/docs/superpowers/plans/2026-05-08-canvas-pr8-8e-wall-overlay.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'Canvas Wall Overlay — PR8 sub-PR 8e implementation plan'
-audience: [engineer, product]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-08
diff --git a/docs/superpowers/plans/2026-05-08-canvas-pr9-legacy-cleanup.md b/docs/superpowers/plans/2026-05-08-canvas-pr9-legacy-cleanup.md
index 7fb83cbdc..123531aec 100644
--- a/docs/superpowers/plans/2026-05-08-canvas-pr9-legacy-cleanup.md
+++ b/docs/superpowers/plans/2026-05-08-canvas-pr9-legacy-cleanup.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Canvas Migration PR9 — Legacy component cleanup
-audience: [engineer]
+audience: human
 category: implementation
-status: delivered
+status: active
 last-reviewed: 2026-05-08
 related:
   - docs/superpowers/specs/2026-05-04-canvas-migration-design.md
diff --git a/docs/superpowers/plans/2026-05-08-improvement-project-v1.md b/docs/superpowers/plans/2026-05-08-improvement-project-v1.md
index a267a28c2..2918f42e5 100644
--- a/docs/superpowers/plans/2026-05-08-improvement-project-v1.md
+++ b/docs/superpowers/plans/2026-05-08-improvement-project-v1.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Improvement Project V1 — Implementation Plan (SUPERSEDED 2026-05-09)
-audience: [engineer]
+audience: human
 category: implementation
 status: superseded
 superseded-by: docs/superpowers/specs/2026-05-09-response-path-system-v1-design.md
diff --git a/docs/superpowers/plans/2026-05-09-response-path-system-v1.md b/docs/superpowers/plans/2026-05-09-response-path-system-v1.md
index 2585e4d2a..e718b418a 100644
--- a/docs/superpowers/plans/2026-05-09-response-path-system-v1.md
+++ b/docs/superpowers/plans/2026-05-09-response-path-system-v1.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Response Path System V1 — Implementation Plan
-audience: [engineer]
+audience: human
 category: implementation
-status: delivered
+status: active
 last-reviewed: 2026-05-13
 related:
   - docs/superpowers/specs/2026-05-09-response-path-system-v1-design.md
diff --git a/docs/superpowers/plans/2026-05-13-canvas-viewport-8f-followups.md b/docs/superpowers/plans/2026-05-13-canvas-viewport-8f-followups.md
index 1cb6400c4..c06df8f84 100644
--- a/docs/superpowers/plans/2026-05-13-canvas-viewport-8f-followups.md
+++ b/docs/superpowers/plans/2026-05-13-canvas-viewport-8f-followups.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Canvas Viewport 8f — Followup Workstream Implementation Plan
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-13
diff --git a/docs/superpowers/plans/2026-05-13-canvas-viewport-architecture.md b/docs/superpowers/plans/2026-05-13-canvas-viewport-architecture.md
index 492eb3a99..0584d4621 100644
--- a/docs/superpowers/plans/2026-05-13-canvas-viewport-architecture.md
+++ b/docs/superpowers/plans/2026-05-13-canvas-viewport-architecture.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Canvas Viewport Architecture (8f) — Implementation Plan
-audience: [engineer]
+audience: human
 category: implementation
-status: delivered
+status: active
 last-reviewed: 2026-05-13
 related:
   - docs/superpowers/specs/2026-05-13-canvas-viewport-architecture-design.md
diff --git a/docs/superpowers/plans/2026-05-14-projects-tab-foundation.md b/docs/superpowers/plans/2026-05-14-projects-tab-foundation.md
index fd683dfa1..e6eba3e49 100644
--- a/docs/superpowers/plans/2026-05-14-projects-tab-foundation.md
+++ b/docs/superpowers/plans/2026-05-14-projects-tab-foundation.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Projects Tab + IP Detail Page Foundation — Implementation Plan
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-14
diff --git a/docs/superpowers/plans/2026-05-14-projects-tab-launchpad-cascade.md b/docs/superpowers/plans/2026-05-14-projects-tab-launchpad-cascade.md
index a0d9a12f8..fc8271ba8 100644
--- a/docs/superpowers/plans/2026-05-14-projects-tab-launchpad-cascade.md
+++ b/docs/superpowers/plans/2026-05-14-projects-tab-launchpad-cascade.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Projects Tab Plan 2 — Active-IP Launchpad + Context Cascade
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-15
diff --git a/docs/superpowers/plans/2026-05-14-projects-tab-report.md b/docs/superpowers/plans/2026-05-14-projects-tab-report.md
index 1a9a955bb..5211fb7e3 100644
--- a/docs/superpowers/plans/2026-05-14-projects-tab-report.md
+++ b/docs/superpowers/plans/2026-05-14-projects-tab-report.md
@@ -1,6 +1,8 @@
 ---
+tier: ephemeral
+purpose: build
 title: Projects Tab Plan 4 — Report Tab IP-Scoped V1
-audience: [engineer]
+audience: human
 category: implementation
 status: active
 last-reviewed: 2026-05-15
diff --git a/docs/superpowers/plans/2026-05-14-projects-tab-team-rail.md b/docs/superpowers/plans/2026-05-14-projects-tab-team-rail.md
index 878e3f725..9d2f7ff44 100644
--- a/docs/superpowers/plans/2026-05-14-projects-tab-team-rail.md
+++ b/docs/superpowers/plans/2026-05-14-projects-tab-team-rail.md
@@ -1,8 +1,10 @@
 ---
+tier: ephemeral
+purpose: build
 title: Projects Tab Plan 3 — Team Workspace Right Rail V1
-audience: [engineer]
+audience: human
 category: implementation
-status: review
+status: draft
 last-reviewed: 2026-05-15
 related:
   - docs/superpowers/specs/2026-05-14-projects-tab-design.md
diff --git a/docs/superpowers/plans/2026-05-16-pr-wv1-1-project-membership.md b/docs/superpowers/plans/2026-05-16-pr-wv1-1-project-membership.md
index b6393902c..6ab4bb4a7 100644
--- a/docs/superpowers/plans/2026-05-16-pr-wv1-1-project-membership.md
+++ b/docs/superpowers/plans/2026-05-16-pr-wv1-1-project-membership.md
@@ -1,4 +1,6 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'PR-WV1-1 — Project Membership Foundation (bite-sized plan)'
 status: draft
 last-reviewed: 2026-05-16
diff --git a/docs/superpowers/plans/2026-05-16-wedge-implementation.md b/docs/superpowers/plans/2026-05-16-wedge-implementation.md
index e354337ce..ffbcbc542 100644
--- a/docs/superpowers/plans/2026-05-16-wedge-implementation.md
+++ b/docs/superpowers/plans/2026-05-16-wedge-implementation.md
@@ -1,4 +1,6 @@
 ---
+tier: ephemeral
+purpose: build
 title: 'Wedge V1 Implementation Plan — Master Sequencer'
 status: draft
 last-reviewed: 2026-05-16
diff --git a/docs/superpowers/specs/2026-03-19-ai-action-tools-design.md b/docs/superpowers/specs/2026-03-19-ai-action-tools-design.md
index 2edf701d8..f8e9e4447 100644
--- a/docs/superpowers/specs/2026-03-19-ai-action-tools-design.md
+++ b/docs/superpowers/specs/2026-03-19-ai-action-tools-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: AI Action Tools Design Spec
-audience: [developer, analyst]
+audience: human
 category: architecture
 status: active
 related: [ai-tools, coscout, function-calling, action-proposal]
diff --git a/docs/superpowers/specs/2026-03-24-adr049-evaluation-report.md b/docs/superpowers/specs/2026-03-24-adr049-evaluation-report.md
index 66c8ba5a0..3fdf517d9 100644
--- a/docs/superpowers/specs/2026-03-24-adr049-evaluation-report.md
+++ b/docs/superpowers/specs/2026-03-24-adr049-evaluation-report.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: design
 title: 'Evaluation Report: PR #33 — ADR-049 CoScout Context & Process Memory'
 status: active
 ---
diff --git a/docs/superpowers/specs/2026-03-30-holistic-evaluation-vqi.md b/docs/superpowers/specs/2026-03-30-holistic-evaluation-vqi.md
index d142a11ab..bf92d4bdb 100644
--- a/docs/superpowers/specs/2026-03-30-holistic-evaluation-vqi.md
+++ b/docs/superpowers/specs/2026-03-30-holistic-evaluation-vqi.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: 'Holistic Evaluation: VariScout Quality Index (VQI)'
-audience: [analyst, engineer]
+audience: human
 category: reference
 status: active
 related: [evaluation, ai-experience, maturity-model, competitive-analysis, architecture]
diff --git a/docs/superpowers/specs/2026-03-31-vqi-rescore-design.md b/docs/superpowers/specs/2026-03-31-vqi-rescore-design.md
index b90491075..a19e10a1d 100644
--- a/docs/superpowers/specs/2026-03-31-vqi-rescore-design.md
+++ b/docs/superpowers/specs/2026-03-31-vqi-rescore-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: 'VQI Re-Evaluation: Post-ADR-053 Scoring + Activation Improvements'
-audience: [analyst, engineer]
+audience: human
 category: reference
 status: active
 related: [evaluation, ai-experience, question-driven-eda, coscout, prompt-engineering]
diff --git a/docs/superpowers/specs/2026-04-02-coscout-intelligence-architecture-design.md b/docs/superpowers/specs/2026-04-02-coscout-intelligence-architecture-design.md
index 4d5ab0b5d..e2803d11a 100644
--- a/docs/superpowers/specs/2026-04-02-coscout-intelligence-architecture-design.md
+++ b/docs/superpowers/specs/2026-04-02-coscout-intelligence-architecture-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'CoScout Intelligence Architecture'
 date: 2026-04-02
-status: in-progress
-audience: [developer, architect]
+status: active
+audience: human
 category: architecture
 related:
   [
diff --git a/docs/superpowers/specs/2026-04-02-deployment-lifecycle-design.md b/docs/superpowers/specs/2026-04-02-deployment-lifecycle-design.md
index fd296c3aa..d40a0f695 100644
--- a/docs/superpowers/specs/2026-04-02-deployment-lifecycle-design.md
+++ b/docs/superpowers/specs/2026-04-02-deployment-lifecycle-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Deployment Lifecycle & Self-Service Update System
-audience: [developer, architect]
+audience: human
 category: architecture
 status: draft
 related: [deployment, ci-cd, azure-marketplace, managed-application, update-mechanism]
diff --git a/docs/superpowers/specs/2026-04-02-knowledge-tab-design.md b/docs/superpowers/specs/2026-04-02-knowledge-tab-design.md
index 67f8fddfb..f3866f1ff 100644
--- a/docs/superpowers/specs/2026-04-02-knowledge-tab-design.md
+++ b/docs/superpowers/specs/2026-04-02-knowledge-tab-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Knowledge Tab & CoScout Inline Citation Preview'
 date: 2026-04-02
 status: draft
-audience: [developer, designer]
+audience: human
 category: architecture
 related: [adr-060, adr-049, adr-057, knowledge-base, coscout, pi-panel]
 ---
diff --git a/docs/superpowers/specs/2026-04-03-investigation-workspace-reframing-design.md b/docs/superpowers/specs/2026-04-03-investigation-workspace-reframing-design.md
index 502cc0255..32d68828c 100644
--- a/docs/superpowers/specs/2026-04-03-investigation-workspace-reframing-design.md
+++ b/docs/superpowers/specs/2026-04-03-investigation-workspace-reframing-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Investigation Workspace Reframing — Problem Statement, Suspected Causes, and the EDA Spine'
-audience: [analyst, engineer]
+audience: human
 category: architecture
-status: in-progress
+status: active
 related:
   [
     investigation,
diff --git a/docs/superpowers/specs/2026-04-03-standard-anova-metrics-design.md b/docs/superpowers/specs/2026-04-03-standard-anova-metrics-design.md
index 403d14d0c..c311ff0ac 100644
--- a/docs/superpowers/specs/2026-04-03-standard-anova-metrics-design.md
+++ b/docs/superpowers/specs/2026-04-03-standard-anova-metrics-design.md
@@ -1,4 +1,6 @@
 ---
+tier: living
+purpose: design
 title: Standard ANOVA Metrics & Category Total SS Removal
 status: draft
 date: 2026-04-03
diff --git a/docs/superpowers/specs/2026-04-04-investigation-spine-design.md b/docs/superpowers/specs/2026-04-04-investigation-spine-design.md
index 2c2a11ca0..e3bf73f5e 100644
--- a/docs/superpowers/specs/2026-04-04-investigation-spine-design.md
+++ b/docs/superpowers/specs/2026-04-04-investigation-spine-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Investigation Spine — Three Threads, Five Sentences, One Story'
-audience: [analyst, engineer]
+audience: human
 category: architecture
-status: delivered
+status: active
 related:
   [
     investigation,
diff --git a/docs/superpowers/specs/2026-04-04-zustand-direct-store-access-design.md b/docs/superpowers/specs/2026-04-04-zustand-direct-store-access-design.md
index 790e06024..5ec664912 100644
--- a/docs/superpowers/specs/2026-04-04-zustand-direct-store-access-design.md
+++ b/docs/superpowers/specs/2026-04-04-zustand-direct-store-access-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Zustand Direct Store Access — Full DataContext Replacement'
-status: delivered
+status: active
 date: 2026-04-04
-audience: [developer, architect]
+audience: human
 category: architecture
 related: [state-management, zustand, DDD, CQRS, persistence]
 ---
diff --git a/docs/superpowers/specs/2026-04-04-zustand-first-state-architecture-design.md b/docs/superpowers/specs/2026-04-04-zustand-first-state-architecture-design.md
index 82e925cf0..822ce6405 100644
--- a/docs/superpowers/specs/2026-04-04-zustand-first-state-architecture-design.md
+++ b/docs/superpowers/specs/2026-04-04-zustand-first-state-architecture-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Zustand-First State Architecture'
-status: delivered
+status: active
 date: 2026-04-04
-audience: [developer, architect]
+audience: human
 category: architecture
 related: [state-management, zustand, persistence, collaboration, DDD]
 ---
diff --git a/docs/superpowers/specs/2026-04-05-evidence-map-design.md b/docs/superpowers/specs/2026-04-05-evidence-map-design.md
index 996df7fad..3decc662f 100644
--- a/docs/superpowers/specs/2026-04-05-evidence-map-design.md
+++ b/docs/superpowers/specs/2026-04-05-evidence-map-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Evidence Map — Data Relationship Visualization
-status: delivered
+status: active
 date: 2026-04-05
-audience: [developer, analyst]
+audience: human
 category: architecture
 related: [best-subsets, factor-intelligence, investigation-spine, causal-graph, pop-out-window]
 ---
diff --git a/docs/superpowers/specs/2026-04-05-evidence-map-spine-design.md b/docs/superpowers/specs/2026-04-05-evidence-map-spine-design.md
index 4772f54cf..7dc10ad4d 100644
--- a/docs/superpowers/specs/2026-04-05-evidence-map-spine-design.md
+++ b/docs/superpowers/specs/2026-04-05-evidence-map-spine-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Evidence Map as Analysis Spine
-status: delivered
+status: active
 date: 2026-04-05
-audience: [developer, analyst]
+audience: human
 category: architecture
 related:
   [evidence-map, factor-intelligence, investigation-spine, pi-panel, coscout, probability-plot]
diff --git a/docs/superpowers/specs/2026-04-07-evidence-map-edge-interactions-design.md b/docs/superpowers/specs/2026-04-07-evidence-map-edge-interactions-design.md
index 8365218c7..5d7cdb461 100644
--- a/docs/superpowers/specs/2026-04-07-evidence-map-edge-interactions-design.md
+++ b/docs/superpowers/specs/2026-04-07-evidence-map-edge-interactions-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Evidence Map Edge Interactions — Statistical Edge → Investigation Bridge
-audience: [developer]
+audience: human
 category: architecture
-status: delivered
+status: active
 related: [evidence-map, evidence-map-spine, investigation, coscout, causal-link]
 ---
 
diff --git a/docs/superpowers/specs/2026-04-07-process-flow-analysis-mode-design.md b/docs/superpowers/specs/2026-04-07-process-flow-analysis-mode-design.md
index 021af9578..e60f1342a 100644
--- a/docs/superpowers/specs/2026-04-07-process-flow-analysis-mode-design.md
+++ b/docs/superpowers/specs/2026-04-07-process-flow-analysis-mode-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Process Flow Analysis Mode
-audience: [analyst, engineer]
+audience: human
 category: analysis
 status: draft
 related: [process-flow, yamazumi, output-rate, bottleneck, lead-time, theory-of-constraints]
diff --git a/docs/superpowers/specs/2026-04-16-defect-analysis-mode-design.md b/docs/superpowers/specs/2026-04-16-defect-analysis-mode-design.md
index 7fe91ba2d..54f6bb409 100644
--- a/docs/superpowers/specs/2026-04-16-defect-analysis-mode-design.md
+++ b/docs/superpowers/specs/2026-04-16-defect-analysis-mode-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Defect Analysis Mode
-audience: [engineer, analyst]
+audience: human
 category: analysis
-status: delivered
+status: active
 related: [defect, pareto, aggregation, mode, strategy-pattern, copq]
 ---
 
diff --git a/docs/superpowers/specs/2026-04-16-defect-evidence-map-integration-design.md b/docs/superpowers/specs/2026-04-16-defect-evidence-map-integration-design.md
index 587c140f3..e07a805e6 100644
--- a/docs/superpowers/specs/2026-04-16-defect-evidence-map-integration-design.md
+++ b/docs/superpowers/specs/2026-04-16-defect-evidence-map-integration-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Defect Mode Evidence Map Integration
-audience: [engineer, analyst]
+audience: human
 category: analysis
-status: delivered
+status: active
 related: [defect, evidence-map, regression, cross-type, investigation]
 ---
 
diff --git a/docs/superpowers/specs/2026-04-17-agent-docs-architecture-design.md b/docs/superpowers/specs/2026-04-17-agent-docs-architecture-design.md
index 468a508b4..962d5dd11 100644
--- a/docs/superpowers/specs/2026-04-17-agent-docs-architecture-design.md
+++ b/docs/superpowers/specs/2026-04-17-agent-docs-architecture-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Agent-Facing Documentation Architecture (A++)
-audience: [engineer]
+audience: human
 category: architecture
-status: delivered
+status: active
 related:
   [claude-md, skills, documentation, governance, agent-context, progressive-disclosure, opus-4-7]
 ---
diff --git a/docs/superpowers/specs/2026-04-17-doc-prune-audit-design.md b/docs/superpowers/specs/2026-04-17-doc-prune-audit-design.md
index e66e9b04d..43b479c12 100644
--- a/docs/superpowers/specs/2026-04-17-doc-prune-audit-design.md
+++ b/docs/superpowers/specs/2026-04-17-doc-prune-audit-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: 'Doc Prune — Load-Bearing Audit Design'
-audience: [engineer, architect]
+audience: human
 category: architecture
 status: draft
 related: [agent-docs-architecture, docs-governance, frontmatter-enforcement]
diff --git a/docs/superpowers/specs/2026-04-18-frame-process-map-design.md b/docs/superpowers/specs/2026-04-18-frame-process-map-design.md
index 4b0678244..4f2832f37 100644
--- a/docs/superpowers/specs/2026-04-18-frame-process-map-design.md
+++ b/docs/superpowers/specs/2026-04-18-frame-process-map-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: FRAME Workspace & Visual Process Map
-audience: [engineer, product]
+audience: human
 category: architecture
-status: delivered
+status: active
 related:
   [
     frame-workspace,
diff --git a/docs/superpowers/specs/2026-04-19-investigation-wall-design.md b/docs/superpowers/specs/2026-04-19-investigation-wall-design.md
index 924a82780..95cf98047 100644
--- a/docs/superpowers/specs/2026-04-19-investigation-wall-design.md
+++ b/docs/superpowers/specs/2026-04-19-investigation-wall-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: 'Investigation Wall — River-Roots Design'
-status: delivered
+status: active
 date: 2026-04-19
-audience: [developer, designer, analyst]
+audience: human
 category: architecture
 related:
   [
diff --git a/docs/superpowers/specs/2026-04-24-coscout-voice-input-design.md b/docs/superpowers/specs/2026-04-24-coscout-voice-input-design.md
index 115d1bc27..5ee684769 100644
--- a/docs/superpowers/specs/2026-04-24-coscout-voice-input-design.md
+++ b/docs/superpowers/specs/2026-04-24-coscout-voice-input-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: CoScout Voice Input v1
-audience: [engineer, product]
+audience: human
 category: architecture
 status: draft
 related: [coscout, azure, voice-input, findings, shop-floor, customer-owned-data]
diff --git a/docs/superpowers/specs/2026-04-25-claude-design-ui-concept-extraction.md b/docs/superpowers/specs/2026-04-25-claude-design-ui-concept-extraction.md
index 3b3e37992..4c85e775c 100644
--- a/docs/superpowers/specs/2026-04-25-claude-design-ui-concept-extraction.md
+++ b/docs/superpowers/specs/2026-04-25-claude-design-ui-concept-extraction.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Claude Design UI Concept Extraction
-audience: [designer, engineer, product]
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/superpowers/specs/2026-04-25-process-flow-yamazumi-integration-design.md b/docs/superpowers/specs/2026-04-25-process-flow-yamazumi-integration-design.md
index 9b8433488..9b1e4c836 100644
--- a/docs/superpowers/specs/2026-04-25-process-flow-yamazumi-integration-design.md
+++ b/docs/superpowers/specs/2026-04-25-process-flow-yamazumi-integration-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Process Flow + Yamazumi Integration
-audience: [analyst, engineer]
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/superpowers/specs/2026-04-25-question-driven-eda-2-design.md b/docs/superpowers/specs/2026-04-25-question-driven-eda-2-design.md
index ff56236c5..fa389717b 100644
--- a/docs/superpowers/specs/2026-04-25-question-driven-eda-2-design.md
+++ b/docs/superpowers/specs/2026-04-25-question-driven-eda-2-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Question-Driven EDA 2.0 — Current Understanding, Mechanism Branches, and Signal Cards
-audience: [analyst, engineer, product]
+audience: human
 category: design-spec
-status: in-progress
+status: active
 related:
   [
     eda-mental-model,
diff --git a/docs/superpowers/specs/2026-04-26-agent-review-log-process-hub-design.md b/docs/superpowers/specs/2026-04-26-agent-review-log-process-hub-design.md
index c2c2a7302..b14196015 100644
--- a/docs/superpowers/specs/2026-04-26-agent-review-log-process-hub-design.md
+++ b/docs/superpowers/specs/2026-04-26-agent-review-log-process-hub-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Agent Review Log Profile - Safe Green Throughput in Process Hub
-audience: [product, engineer, quality-manager]
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/superpowers/specs/2026-04-26-evidence-sources-data-profiles-design.md b/docs/superpowers/specs/2026-04-26-evidence-sources-data-profiles-design.md
index 239b55971..11ee85f19 100644
--- a/docs/superpowers/specs/2026-04-26-evidence-sources-data-profiles-design.md
+++ b/docs/superpowers/specs/2026-04-26-evidence-sources-data-profiles-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Evidence Sources And Data Profiles - Process Hub Evidence Workflow
-audience: [product, designer, engineer]
+audience: human
 category: design-spec
-status: in-progress
+status: active
 related:
   [process-hub, evidence-sources, data-profiles, snapshots, customer-owned-data, agent-review-log]
 date: 2026-04-26
diff --git a/docs/superpowers/specs/2026-04-27-phase-3-h1-closure-h2-launch-design.md b/docs/superpowers/specs/2026-04-27-phase-3-h1-closure-h2-launch-design.md
index 208b899ed..1ff3522bf 100644
--- a/docs/superpowers/specs/2026-04-27-phase-3-h1-closure-h2-launch-design.md
+++ b/docs/superpowers/specs/2026-04-27-phase-3-h1-closure-h2-launch-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Phase 3 — H1 Closure + H2 Launch
-audience: engineer
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/superpowers/specs/2026-04-28-production-line-glance-design.md b/docs/superpowers/specs/2026-04-28-production-line-glance-design.md
index eec93ff2f..c1657d86a 100644
--- a/docs/superpowers/specs/2026-04-28-production-line-glance-design.md
+++ b/docs/superpowers/specs/2026-04-28-production-line-glance-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Production-Line-Glance Dashboard — Per-Step Capability with Context-Aware Specs
-audience: [product, designer, engineer, analyst, manager]
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/superpowers/specs/2026-04-28-production-line-glance-surface-wiring-design.md b/docs/superpowers/specs/2026-04-28-production-line-glance-surface-wiring-design.md
index 7ca8913b6..0f3d45ced 100644
--- a/docs/superpowers/specs/2026-04-28-production-line-glance-surface-wiring-design.md
+++ b/docs/superpowers/specs/2026-04-28-production-line-glance-surface-wiring-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Production-Line-Glance Surface Wiring (Plan C) — One Dashboard, Three Surfaces
-audience: [product, designer, engineer, analyst, manager]
+audience: human
 category: design-spec
 status: draft
 related:
diff --git a/docs/superpowers/specs/2026-04-29-consolidated-method-and-surface-overview-design.md b/docs/superpowers/specs/2026-04-29-consolidated-method-and-surface-overview-design.md
index 86c357c29..f1c44f2f0 100644
--- a/docs/superpowers/specs/2026-04-29-consolidated-method-and-surface-overview-design.md
+++ b/docs/superpowers/specs/2026-04-29-consolidated-method-and-surface-overview-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: VariScout — Consolidated Method And Surface Overview
-audience: [product, designer, engineer, analyst]
+audience: human
 category: design-spec
-status: review
+status: draft
 related:
   [
     overview,
diff --git a/docs/superpowers/specs/2026-04-29-investigation-scope-and-drill-semantics-design.md b/docs/superpowers/specs/2026-04-29-investigation-scope-and-drill-semantics-design.md
index 94038d3c2..702271157 100644
--- a/docs/superpowers/specs/2026-04-29-investigation-scope-and-drill-semantics-design.md
+++ b/docs/superpowers/specs/2026-04-29-investigation-scope-and-drill-semantics-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Investigation Scope and Drill Semantics
-audience: [product, designer, engineer, analyst]
+audience: human
 category: design-spec
-status: stable
+status: active
 related:
   [
     process-hub,
diff --git a/docs/superpowers/specs/2026-04-29-multi-level-scout-design.md b/docs/superpowers/specs/2026-04-29-multi-level-scout-design.md
index b42f17227..19be2c414 100644
--- a/docs/superpowers/specs/2026-04-29-multi-level-scout-design.md
+++ b/docs/superpowers/specs/2026-04-29-multi-level-scout-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Multi-level SCOUT — design
-audience: [product, designer, engineer, analyst]
+audience: human
 category: design-spec
-status: delivered
+status: active
 date: 2026-04-29
 last-reviewed: 2026-04-30
 related:
diff --git a/docs/superpowers/specs/2026-05-03-scout-ui-consolidation-design.md b/docs/superpowers/specs/2026-05-03-scout-ui-consolidation-design.md
index 321c17486..d3b58dbe0 100644
--- a/docs/superpowers/specs/2026-05-03-scout-ui-consolidation-design.md
+++ b/docs/superpowers/specs/2026-05-03-scout-ui-consolidation-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Scout UI consolidation — design
-audience: [designer, engineer, product]
+audience: human
 category: design-spec
 status: draft
 date: 2026-05-03
diff --git a/docs/superpowers/specs/2026-05-03-variscout-vision-design.md b/docs/superpowers/specs/2026-05-03-variscout-vision-design.md
index 17053463b..65b8f8356 100644
--- a/docs/superpowers/specs/2026-05-03-variscout-vision-design.md
+++ b/docs/superpowers/specs/2026-05-03-variscout-vision-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: VariScout — Product Vision
-audience: [product, engineer, designer, analyst, business]
+audience: human
 category: strategy
-status: accepted
+status: active
 last-reviewed: 2026-05-03
 supersedes:
   - docs/archive/specs/2026-04-27-process-learning-operating-model-design.md
diff --git a/docs/superpowers/specs/2026-05-04-canvas-migration-design.md b/docs/superpowers/specs/2026-05-04-canvas-migration-design.md
index 531a873f0..731ad7926 100644
--- a/docs/superpowers/specs/2026-05-04-canvas-migration-design.md
+++ b/docs/superpowers/specs/2026-05-04-canvas-migration-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Canvas Migration Strategy — strangler-pattern phased delivery + three-layer state separation
-audience: [product, engineer, designer]
+audience: human
 category: design-spec
 status: active
 last-reviewed: 2026-05-05
diff --git a/docs/superpowers/specs/2026-05-04-manual-canvas-authoring-design.md b/docs/superpowers/specs/2026-05-04-manual-canvas-authoring-design.md
index 59764db20..c9c8f36ec 100644
--- a/docs/superpowers/specs/2026-05-04-manual-canvas-authoring-design.md
+++ b/docs/superpowers/specs/2026-05-04-manual-canvas-authoring-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Manual Canvas Authoring — Spec 2 of 5 canvas-detail specs
-audience: [product, engineer, designer]
+audience: human
 category: design-spec
 status: active
 last-reviewed: 2026-05-04
diff --git a/docs/superpowers/specs/2026-05-06-data-flow-foundation-design.md b/docs/superpowers/specs/2026-05-06-data-flow-foundation-design.md
index 5eae23d6e..0ec62d1ac 100644
--- a/docs/superpowers/specs/2026-05-06-data-flow-foundation-design.md
+++ b/docs/superpowers/specs/2026-05-06-data-flow-foundation-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Data-Flow Foundation — repository pattern + normalized persistence + uniform action layer
-audience: [product, engineer, designer]
+audience: human
 category: design-spec
-status: delivered
+status: active
 last-reviewed: 2026-05-06
 related:
   - docs/superpowers/specs/2026-05-03-variscout-vision-design.md
diff --git a/docs/superpowers/specs/2026-05-07-canvas-hypothesis-arrow-drawing-design.md b/docs/superpowers/specs/2026-05-07-canvas-hypothesis-arrow-drawing-design.md
index 62507a154..c85a91b66 100644
--- a/docs/superpowers/specs/2026-05-07-canvas-hypothesis-arrow-drawing-design.md
+++ b/docs/superpowers/specs/2026-05-07-canvas-hypothesis-arrow-drawing-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Canvas Hypothesis-Arrow Drawing — PR8 sub-PR 8d design (Spec 4 extension)
-audience: [engineer, product, designer]
+audience: human
 category: design-spec
 status: active
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/specs/2026-05-07-data-flow-foundation-f4-three-layer-state-design.md b/docs/superpowers/specs/2026-05-07-data-flow-foundation-f4-three-layer-state-design.md
index 914d50c0e..464201224 100644
--- a/docs/superpowers/specs/2026-05-07-data-flow-foundation-f4-three-layer-state-design.md
+++ b/docs/superpowers/specs/2026-05-07-data-flow-foundation-f4-three-layer-state-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Data-Flow Foundation F4 — Document / Annotation / View state codification
-audience: [product, engineer]
+audience: human
 category: design-spec
-status: delivered
+status: active
 last-reviewed: 2026-05-07
 related:
   - docs/superpowers/specs/2026-05-06-data-flow-foundation-design.md
diff --git a/docs/superpowers/specs/2026-05-07-security-hardening-design.md b/docs/superpowers/specs/2026-05-07-security-hardening-design.md
index 1b1c63de3..95ea8a961 100644
--- a/docs/superpowers/specs/2026-05-07-security-hardening-design.md
+++ b/docs/superpowers/specs/2026-05-07-security-hardening-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Security Hardening Concept — Azure VariScout (hub authz + SAS scope + login guardrails)
-audience: [product, engineer]
+audience: human
 category: design-spec
 status: draft
 last-reviewed: 2026-05-07
diff --git a/docs/superpowers/specs/2026-05-08-canvas-wall-overlay-design.md b/docs/superpowers/specs/2026-05-08-canvas-wall-overlay-design.md
index 41df93b27..91d12c938 100644
--- a/docs/superpowers/specs/2026-05-08-canvas-wall-overlay-design.md
+++ b/docs/superpowers/specs/2026-05-08-canvas-wall-overlay-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Canvas Wall Overlay — PR8 sub-PR 8e design (Spec 4 extension)
-audience: [engineer, product, designer]
+audience: human
 category: design-spec
-status: delivered
+status: active
 last-reviewed: 2026-05-08
 related:
   - docs/superpowers/specs/2026-05-03-variscout-vision-design.md
diff --git a/docs/superpowers/specs/2026-05-08-improvement-project-v1-design.md b/docs/superpowers/specs/2026-05-08-improvement-project-v1-design.md
index b2b5cb790..b8929e387 100644
--- a/docs/superpowers/specs/2026-05-08-improvement-project-v1-design.md
+++ b/docs/superpowers/specs/2026-05-08-improvement-project-v1-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Improvement Project V1 — Design (SUPERSEDED 2026-05-09)
-audience: [product, engineer, designer]
+audience: human
 category: design-spec
 status: superseded
 superseded-by: docs/superpowers/specs/2026-05-09-response-path-system-v1-design.md
diff --git a/docs/superpowers/specs/2026-05-09-response-path-system-v1-design.md b/docs/superpowers/specs/2026-05-09-response-path-system-v1-design.md
index d6ef7847f..3edbd90e8 100644
--- a/docs/superpowers/specs/2026-05-09-response-path-system-v1-design.md
+++ b/docs/superpowers/specs/2026-05-09-response-path-system-v1-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Response Path System V1 — Design
-audience: [product, engineer, designer]
+audience: human
 category: design-spec
-status: delivered
+status: active
 last-reviewed: 2026-05-13
 supersedes:
   - docs/superpowers/specs/2026-05-08-improvement-project-v1-design.md
diff --git a/docs/superpowers/specs/2026-05-13-canvas-viewport-architecture-design.md b/docs/superpowers/specs/2026-05-13-canvas-viewport-architecture-design.md
index 10d4c3e50..17dcd25d9 100644
--- a/docs/superpowers/specs/2026-05-13-canvas-viewport-architecture-design.md
+++ b/docs/superpowers/specs/2026-05-13-canvas-viewport-architecture-design.md
@@ -1,8 +1,10 @@
 ---
+tier: living
+purpose: design
 title: Canvas Viewport Architecture — Design (8f)
-audience: [product, engineer, designer]
+audience: human
 category: design-spec
-status: delivered
+status: active
 last-reviewed: 2026-05-13
 related:
   - docs/superpowers/specs/2026-05-03-variscout-vision-design.md
diff --git a/docs/superpowers/specs/2026-05-14-projects-tab-design.md b/docs/superpowers/specs/2026-05-14-projects-tab-design.md
index 6dced1cab..ae83fe0b3 100644
--- a/docs/superpowers/specs/2026-05-14-projects-tab-design.md
+++ b/docs/superpowers/specs/2026-05-14-projects-tab-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Projects tab + IP Detail (Lifecycle) page + Home active-IP launchpad
-audience: [product, engineer, designer]
+audience: human
 category: design-spec
 status: draft
 last-reviewed: 2026-05-14
diff --git a/docs/superpowers/specs/2026-05-14-variscout-coherence-design.md b/docs/superpowers/specs/2026-05-14-variscout-coherence-design.md
index 2bce2abad..ed8e2b7a0 100644
--- a/docs/superpowers/specs/2026-05-14-variscout-coherence-design.md
+++ b/docs/superpowers/specs/2026-05-14-variscout-coherence-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: VariScout Coherence — Persona + Surface + Vocabulary Design
-audience: [product, designer, engineer, analyst]
+audience: human
 category: design-spec
 status: draft
 last-reviewed: 2026-05-14
diff --git a/docs/superpowers/specs/2026-05-16-docs-strategy-design.md b/docs/superpowers/specs/2026-05-16-docs-strategy-design.md
new file mode 100644
index 000000000..5b408de6a
--- /dev/null
+++ b/docs/superpowers/specs/2026-05-16-docs-strategy-design.md
@@ -0,0 +1,521 @@
+---
+title: 'VariScout Docs Strategy 2026 — Full Design'
+status: active
+date: 2026-05-16
+purpose: design
+tier: living
+audience: both
+topic: [docs-strategy, meta, schema]
+related: [2026-05-16-docs-strategy-memo, adr-083-eight-purpose-doc-taxonomy]
+---
+
+# VariScout Docs Strategy 2026 — Full Design
+
+> 🟡 **Active build via [`PR #184`](https://github.com/jukka-matti/variscout/pull/184)** | 🔄 **Last material edits 2026-05-17** — §2.7 expanded with: "Core principles (in priority order)" affirmative list (4 principles); "Reader-first banners at the top" (template matrix + Play 2b validator extensions); "Standard entry format" for decision-log; agent-manifests + investigations + memory added to doc types. Per the principle this spec defines: frontmatter alone is insufficient — this banner signals reader-relevant changes since the spec was first shipped.
+
+_1-page CTO memo: [2026-05-16-docs-strategy-memo.md](2026-05-16-docs-strategy-memo.md). Schema change ADR: [ADR-083](../../07-decisions/adr-083-eight-purpose-doc-taxonomy.md)._
+
+---
+
+## 1. Context
+
+VariScout's documentation system reflects two years of disciplined authoring — 521 markdown files, 74 ADRs, 50 active specs, validation tooling, frontmatter SSOT, atomic-notes memory, nested CLAUDE.md hierarchy. The investment pays off: the recent wedge pivot was managed entirely through doc artifacts.
+
+But the system is **scaling its own friction**:
+
+- `MEMORY.md` is **over its 24.4KB size limit** (36KB; Claude loads partially — agent doesn't know what it doesn't know).
+- `decision-log.md` is **170KB** and growing as the re-litigation gate; queryable only by reading the whole thing.
+- **Phase C audit (~500 docs)** is pending after the wedge pivot — drift fix is scheduled as a project, not as a continuous loop.
+- **Multi-canonical claimants**: vision lives in 5+ places (positioning, business-bible, OVERVIEW, USER-JOURNEYS, root CLAUDE.md, methodology); decisions live in 3 places (ADRs, decision-log, MEMORY).
+- **Metadata over-engineered**: 22 STATUS × 18 CATEGORY × 14 AUDIENCE × 11 numbered folders.
+- **Lifecycle under-engineered**: no freshness signal, no read telemetry, no spec ↔ implementation drift sensor.
+- **Agent Context (AX-dev surface) under-engineered**: 2 skills + 6 globally-loaded rules (path-scoping broken upstream). By 2026 conventions, a project this size has 15+ task-typed skills.
+
+The fix is not "more documentation infrastructure" — it's **less, but living**: simpler metadata, atomic decomposition of bloat sources, **single-source-of-truth discipline by doc type** (§2.7), agent-queryable retrieval, continuous freshness, expanded skill inventory.
+
+**Why doc-type discipline specifically**: a real failure mode surfaced during execution — a mid-flight design change to the wedge spec was captured as a _separate_ `*-amendment-design.md` spec file instead of editing the canonical spec in place. A new session reading only the wedge spec saw stale content and went down a wrong path. The amendment file was discoverable in principle (it was in `docs/superpowers/specs/`) but invisible in practice (no link from the canonical it amended). Convention alone doesn't prevent this — it has to be mechanically enforced (§2.7 + revised Play 2b).
+
+---
+
+## 2. The Foundation: 8 Purposes × 4 Tiers
+
+### 2.1 The 8 purposes
+
+Every doc serves exactly one purpose. Purpose drives **shape and tooling**.
+
+| #     | Purpose           | Reader does this with it                                                        | Failure mode                                 | Example artifacts                                                                      |
+| ----- | ----------------- | ------------------------------------------------------------------------------- | -------------------------------------------- | -------------------------------------------------------------------------------------- |
+| **A** | **orient**        | "What is VariScout? Why? For whom? At what price?"                              | Stale narrative misleads about current state | positioning, business-bible, methodology, glossary, roadmap, tier-philosophy, OVERVIEW |
+| **B** | **decide**        | "What did we conclude?" (incl. open questions awaiting decision)                | Re-litigation; lost provenance               | 74 ADRs, decision-log, investigations                                                  |
+| **C** | **design**        | "What is/should the product be?" (any scope — UX _and_ AX design)               | Spec ↔ implementation drift                  | journeys, 50 specs, design system, features, use cases, tutorials                      |
+| **D** | **build**         | "How do I implement this _here_?" (procedure)                                   | Agent uses outdated procedure                | `.claude/skills/`, plans, procedure sections in package CLAUDE.mds, `10-development/`  |
+| **F** | **system**        | "How is the system architected?"                                                | Diagrams describe old architecture           | DATA-FLOW, `05-technical/`, dependency maps                                            |
+| **G** | **constrain**     | "What must I never do?" (invariants)                                            | Constraint ignored; not enforced             | `.claude/rules/`, constitution, ADR-073, P5/P7-style principles                        |
+| **H** | **agent-context** | "I'm an agent — what do I need to know right now?" (the AX-dev runtime surface) | Context bloat; wrong information loaded      | root + 9 nested CLAUDE.md, AGENTS.md, llms.txt, MEMORY.md, skills (router half)        |
+| **I** | **remember**      | "How did we get here? Why was X decided?" (provenance)                          | Lost institutional knowledge                 | decision-log Replayed Decisions, archived specs, baseline snapshots, transcripts       |
+
+### 2.2 The 4 tiers (velocity)
+
+Tier drives **lifecycle and maintenance cadence**.
+
+| Tier          | Volatility                              | Examples                                          | Maintenance                      |
+| ------------- | --------------------------------------- | ------------------------------------------------- | -------------------------------- |
+| **stable**    | Years-stable                            | vision, positioning, methodology, glossary        | Quarterly human review           |
+| **living**    | Months-stable, amend-not-rewrite        | ADRs, specs, journeys, design system, features    | Steward-loop drift checks weekly |
+| **ephemeral** | Days-to-weeks, lifecycle = until-merged | active plans, open investigations                 | Auto-archive on close            |
+| **card**      | Atomic, append-only                     | decision cards, memory cards, investigation cards | Continuous (Steward + telemetry) |
+
+### 2.3 Two-axis schema
+
+```yaml
+purpose: orient | decide | design | build | system | constrain | agent-context | remember
+tier: stable | living | ephemeral | card
+topic: [<kebab-tags>] # ux, ax, coscout, canvas, stats, etc. — free-form
+status: draft | active | superseded | archived
+audience: human | agent | both
+# operational
+last-verified: YYYY-MM-DD
+verified-against-commit: <sha>
+supersedes: [<id>]
+related: [<id>]
+```
+
+**Collapse from current schema**:
+
+- 22 STATUS enums → **4** (`draft, active, superseded, archived`). Most of the 22 are aspirational distinctions never used.
+- 18 CATEGORY enums → **free-form `topic` tags**. Enums lock you into a 2024 ontology; tags evolve.
+- 14 AUDIENCE enums → **3** (`human, agent, both`). You have one external audience: you + your subagents.
+- 11 numbered folders → **3 velocity-tier folders + cards/** (see §2.5).
+- **New fields**: `purpose` (the foundation), `tier` (velocity), `last-verified`, `verified-against-commit` (drift sensors).
+
+### 2.4 UX, AX, CoScout — handled as topic tags
+
+Rather than splitting these into new purposes, they ride on `topic`:
+
+- UX-flavored Design: `topic: [ux, journey]`, `topic: [ux, design-system]`, `topic: [ux, feature]`
+- AX-flavored Design: `topic: [ax, agent-dispatch]`, `topic: [ax, coscout]`, `topic: [ax, subagent-workflow]`
+- CoScout-specific (cross-purpose): `topic: [coscout, prompts]`, `topic: [coscout, knowledge]`, `topic: [coscout, safety]`
+
+Query: `pnpm docs:find --purpose=design --topic=ax` returns AX-design docs only. Or `--topic=coscout` returns everything CoScout-related across all purposes.
+
+### 2.5 Folder restructure
+
+```
+docs/
+  stable/                       # purpose-mixed, all tier=stable: vision, methodology, glossary, baseline
+  living/                       # purpose-mixed, all tier=living
+    orient/                    # roadmap, tier-philosophy
+    decide/                    # ADRs (formerly 07-decisions/)
+    design/                    # specs, journeys, design system, features
+    build/                     # 10-development, plans (until merged)
+    system/                    # 05-technical, DATA-FLOW
+    agent-context/             # AGENTS.md, llms.txt, root CLAUDE.md (link only — root stays at repo root)
+  ephemeral/                    # active plans, open investigations
+    plans/                     # active plans only; closed → cards/
+    investigations/            # open observations only; closed → cards/
+    transcripts/               # selected plan promotions (Play 5)
+  cards/                        # atomic units, queryable
+    decisions/                 # one card per decision (from decision-log + ADRs after maturation)
+    memory/                    # atomic feedback/user/project memories (mirrored from ~/.claude/memory)
+    investigations/            # closed-but-not-decided observations
+  archive/                      # superseded, retained for git history readers
+```
+
+Nested package `CLAUDE.md` files stay in `packages/*/` and `apps/*/` (progressive disclosure is correct; don't move).
+
+### 2.6 Per-purpose treatment
+
+| Purpose           | Shape                                                                                               | Tooling                                                                                                                                | Lifecycle treatment                                                                        |
+| ----------------- | --------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------ |
+| **orient**        | ≤6 canonical narrative docs                                                                         | Addendum threads; pinned amendments                                                                                                    | Stable; manually reviewed quarterly                                                        |
+| **decide**        | Atomic cards                                                                                        | `pnpm docs:find/get/related`; supersession chains                                                                                      | Card-tier; continuous via Steward                                                          |
+| **design**        | Living specs with `implements:` ADR back-link; **edit in place** (no `*-amendment-*.md` side files) | Spec ↔ impl drift sensor (cited symbols → AST check, Playwright selector check for UX-Design); `delivered-by:` frontmatter links to PR | `draft → active → delivered → superseded → archived` (body always reflects current intent) |
+| **build**         | Skills (procedures) + ephemeral plans + package CLAUDE.md procedure sections                        | `.claude/skills/`; subagent-driven-development                                                                                         | Skills: living; Plans: ephemeral; patterns in nested CLAUDE.md                             |
+| **system**        | Auto-generated where possible (deps, exports), narrated where not                                   | `pnpm docs:gen-arch` from package.json + tsconfig                                                                                      | Living; auto-regenerate on push                                                            |
+| **constrain**     | Rules + linting + tests (NOT prose)                                                                 | ESLint + pre-commit + denylist scripts                                                                                                 | Stable; rules removed only via ADR                                                         |
+| **agent-context** | Progressive disclosure: root → nested → skills → cards                                              | Skill auto-load by description; CLAUDE.md size budget; routing skill replaces static catalog                                           | Living; weekly size check; skill inventory growth                                          |
+| **remember**      | Chronological cards + selected transcript promotions; **append-only**, never edit prior entries     | `docs/cards/` for B+I; `docs/ephemeral/transcripts/` for selected plans                                                                | Append-only; new entry supersedes old when reality changes                                 |
+
+### 2.7 Doc discipline: SSoT by doc type
+
+The 8-purpose × 4-tier foundation defines WHERE docs live. This section defines HOW they update — which is just as important for keeping the corpus coherent. **Without this discipline, the metadata gains are erased by amendment-of-amendment drift.**
+
+#### Core principles (in priority order)
+
+1. **Doc-type discipline.** Different doc types serve different jobs and update differently. Design specs edit-in-place; ADRs append amendment blocks; decision-log appends; generated docs auto-rebuild; plan files stay ephemeral. (Detail in the doc-types table below.)
+
+2. **One canonical home per concept.** Every concept has exactly ONE authoritative doc that owns it. Other docs link rather than restate. When the canonical changes, all linkers point to the same updated truth — no fork in interpretation. Play 4 enforces operationally: audit the corpus for multi-canonical claimants, consolidate, mark losers `superseded` with redirect banners.
+
+3. **Reader-first banners.** Frontmatter is machine-readable; banners are human/agent-readable. When a doc's state diverges from "body is current truth" — superseded, materially edited recently, delivered, ADR with amendments below — say so in a banner at the top. First thing the reader sees.
+
+4. **Decision-log as temporal index.** Every canonical edit produces a decision-log entry in a standard format (see below). Canonical docs are "as of now"; the log is "what changed when and why" over the canonical set.
+
+Conflating these or treating any one as optional creates the kind of drift the 2026-05-17 wedge-amendment incident surfaced — a side-amendment spec violated principles 1, 2, and 3 simultaneously.
+
+#### Doc types at a glance
+
+Different doc types are NOT updated the same way:
+
+- **Design specs are current-state SoT.** The body at any moment IS the truth. Edit in place when intent changes.
+- **ADRs are point-in-time decisions.** Preserve original reasoning. Append `## Amendment — YYYY-MM-DD` blocks for clarifications. New ADR for fundamental reframings.
+- **Decision-log is chronological.** Append-only. Never edit prior entries. New entry supersedes old.
+- **Generated docs are mechanical projections of source.** Auto-rebuilt by scripts. Never hand-edited. `.prettierignore` excludes.
+- **Plan files are ephemeral session transcripts.** Live in `~/.claude/plans/`, not in repo. Selective promotion to `docs/ephemeral/transcripts/` for landmark sessions.
+- **Investigations are pre-decision observations.** Open entries editable while open. `[RESOLVED]` entries immutable. Closed investigations graduate to cards (Play 2a).
+- **Memory holds cross-session durable facts for Claude.** Topic files in `~/.claude/projects/.../memory/` edit in place. Index `MEMORY.md` is one-line entries ≤200 chars (over-budget triggers truncated load).
+- **Agent manifests are agent-loaded context** (root `CLAUDE.md`, `AGENTS.md`, `docs/llms.txt`, nested package `CLAUDE.md`). Edit in place. Per-file CLAUDE.md size budget enforced (`scripts/check-claude-md-size.sh`). These are the "first read" surface for every agent dispatch — keep them tight.
+
+#### Anti-patterns (mechanically forbidden)
+
+| Anti-pattern                                                                     | Why wrong                                                                            | Right pattern                                                                                     | Enforcement                                                            |
+| -------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------- |
+| `*-amendment-design.md`, `*-revision-design.md`, `*-update-design.md` side files | Orphan amendments drift from canonical; readers must mentally merge body + side file | Edit canonical spec in place + decision-log entry                                                 | Validator HARD-FAILs on filenames matching these patterns              |
+| Editing ADR body after `status: accepted`                                        | Loses original decision context                                                      | `## Amendment — YYYY-MM-DD` block at bottom OR new ADR                                            | Validator WARNs on diff to ADR body without amendment-block syntax     |
+| Editing prior decision-log entries                                               | Breaks chronological record                                                          | New entry that supersedes old (with `[supersedes <prior-entry-date>]` marker)                     | Validator WARNs on git diff to lines older than 7 days in decision-log |
+| Hand-editing `*-generated.md` files                                              | Out-of-sync with source; next regen wipes edits                                      | Update the generator; regenerate                                                                  | `.prettierignore` + pre-push staleness check                           |
+| Putting active-decision content only in plan files                               | Plan files live in `~/.claude/plans/`, not in repo; future sessions don't see them   | Land decisions in canonical home (ADR / spec / decision-log) before merging the PR they came from | Reviewer convention; investigations.md tracks gaps                     |
+| Renaming or moving canonical docs without a `supersedes:` chain                  | Breaks inbound links + obscures lineage                                              | `git mv` + add `supersedes:` to the new path's frontmatter + `redirect-to:` banner at old         | Validator checks `supersedes:` chain integrity                         |
+
+#### Decision-log as temporal index
+
+Every change to a canonical doc (spec body edit, ADR amendment, supersession) gets a decision-log entry that says:
+
+- **What** changed (which canonical doc + section)
+- **Why** (link to brainstorm transcript, plan, or PR)
+- **Supersession** (what's now stale vs what's still valid)
+- **Commit reference** (so future readers can git archaeology)
+
+This makes the decision-log the chronological backbone over the canonical doc set. The canonical docs themselves are always "as of now"; the log tells you "what changed when and why".
+
+**Standard entry format** (Play 2c's `pnpm docs:recent --amendments` parses this shape):
+
+```
+- YYYY-MM-DD — <short title ≤80 chars>. <Edit type>: <doc>#<section> [supersedes <prior>].
+  Why: <one-sentence rationale>.
+  Commit: <sha-short>. PR: #N. Related: [[<id>], [<id>]].
+```
+
+**Required fields**: date, title, edit-type (`spec edit` / `ADR amendment` / `new ADR` / `supersession` / `archived` / etc.), `[supersedes <doc>#<section>]` marker if applicable, Why, Commit.
+**Optional**: PR number, Related links (wikilinks to ADRs / specs / other entries).
+
+**Why this format**: the `[supersedes <doc>#<section>]` marker is greppable + machine-parseable (enables `pnpm docs:recent --supersedes <doc>` queries). The 80-char title cap keeps chronological scan readable. The mandatory Why field prevents "the change is the change" entries with no rationale (a real problem in long-lived logs).
+
+**What NOT to put in decision-log entries**: full rationale paragraphs (those live in the brainstorm transcript or PR description that Why links to); restated spec content (the canonical spec is the SoT — log just notes the edit); duplicated information across entries (one decision = one entry, with future entries citing it via `[supersedes …]`).
+
+#### Edit-in-place mechanics (for design specs)
+
+When editing a canonical spec mid-flight:
+
+1. Make the edit.
+2. Update `last-verified: <today>` frontmatter; bump `verified-against-commit: <sha>` after commit.
+3. If the edit is material (changes intent or supersedes a section), **add a status banner at the top** of the body (after the H1) — see below.
+4. Add decision-log entry citing the section change with a `[supersedes <doc>#<section>]` marker.
+5. Commit with message capturing both the spec edit + the log entry.
+6. (If during active build) flag the change in the build PR description so reviewers see it.
+
+This makes mid-flight changes auditable without polluting the spec body with amendment scar tissue.
+
+#### Reader-first banners at the top
+
+**Frontmatter alone is not enough** — frontmatter is for tooling; readers (humans + agents) need a human-readable banner at the TOP of the body, right after the H1. This is the active signal a fresh-session reader sees first.
+
+Required when: `status: superseded`, `status: archived`, `supersedes:` set, material in-place edit in the last 30 days, ADR with amendments below. Recommended for: `status: delivered` (link the shipping PR + code paths).
+
+The wedge-amendment incident proved why: the wedge spec was amended on 2026-05-16 but never got a banner. A fresh session reading only the wedge spec saw no signal → went down the wrong path. A `🔄 Last material edit 2026-05-16 — §3.1 rewritten (Improve tab retained)` banner at the top would have made the divergence immediately visible.
+
+**Pattern**: status banner at TOP for "you must know this before reading the body"; amendment blocks at BOTTOM (ADRs only) for "history accumulates here". Wikilinks (`[[name]]`) are **passive** at read time and don't substitute for banners — they preserve graph structure, not reading priority.
+
+Concrete banner templates + the full required/recommended matrix live in [`docs/agent-context/doc-discipline.md` §Reader-first banners](../../agent-context/doc-discipline.md). Loaded by `agent-context-quickstart` skill on session start so every subagent gets the templates cold.
+
+**Enforcement (Play 2b validator extensions)**:
+
+- HARD-FAIL: `status: superseded` without `> SUPERSEDED` banner in first 10 lines of body
+- HARD-FAIL: `supersedes: [<id>]` set without banner mentioning what's being superseded
+- WARN: `status: delivered` without delivery banner OR `delivered-by:` frontmatter
+- WARN: `status: active` design spec with `last-verified` >30 days behind HEAD (suggests freshness banner or transition to `delivered`)
+
+#### Spec lifecycle states (explicit)
+
+`status:` transitions for design specs:
+
+| Status       | Meaning                                                                   | Body update pattern                                                             |
+| ------------ | ------------------------------------------------------------------------- | ------------------------------------------------------------------------------- |
+| `draft`      | Designing; not yet building                                               | Edit freely; no `last-verified` discipline yet                                  |
+| `active`     | Designed; build underway                                                  | Edit in place when intent changes; bump `last-verified`                         |
+| `delivered`  | Shipped; `delivered-by: PR #N` set; `code-locations: [...]` if applicable | Body describes current state of shipped feature; edit only when feature evolves |
+| `superseded` | Replaced by another spec via `supersedes:` chain                          | Frozen; redirect banner at top pointing to successor                            |
+| `archived`   | Historical reference only                                                 | Frozen; moved to `docs/archive/`; no longer cited from canonical surfaces       |
+
+Transitions are explicit + auditable (frontmatter diff is the signal).
+
+#### Why this prevents "specs over specs"
+
+The user's concern: superpowers methodology creates plan files + design specs + implementation; over time, artifacts accumulate. The fix isn't fewer artifacts — it's clearer ownership:
+
+- **Plan files**: ephemeral; never become canonical
+- **Design specs**: ONE per concept; edit in place during build; supersede only for material reframings
+- **ADRs**: point-in-time decisions; capture reasoning at time T; rarely rewritten
+- **Decision-log**: chronological index over the canonical set; the temporal "what changed when"
+
+At any moment, the question "what is the current intended state of X?" has exactly ONE answer: the canonical spec for X. No mental merging. No side-spec hunting. No "what supersedes what" archaeology in plain reading.
+
+This is the discipline. Without it, the 8-purpose × 4-tier foundation is just metadata gymnastics. With it, the corpus stays coherent as it scales.
+
+---
+
+## 3. The 8 Engineering Plays
+
+Each play is independently shippable. Sequence is deliberate (P1 unlocks P2; P3 best after P2 for skills-citing-toolbox; P4 best after P2 so consolidation cites cards).
+
+### Play 1 — Velocity-tier restructure + schema simplification (3 days)
+
+**Objective**: Move docs into purpose × tier folders + collapse frontmatter schema.
+
+**Deliverables**:
+
+- New folder hierarchy under `docs/stable/`, `docs/living/<purpose>/`, `docs/ephemeral/`, `docs/cards/`
+- `scripts/docs-frontmatter-schema.mjs` updated: 4 STATUS, free-form topic, 3 AUDIENCE, new `purpose` + `tier` + drift fields
+- `scripts/docs-frontmatter-fix.mjs` extended to backfill `purpose` + `tier` based on inferred mapping
+- All 521 docs `git mv`'d into new locations (history preserved)
+- `docs/llms.txt` updated to reflect new tree
+- Root `CLAUDE.md` "Where to look" section updated
+
+**Exit criteria**: `pnpm docs:health` green; all internal links resolve; `bash scripts/pr-ready-check.sh` green.
+
+### Play 2 — Cards & Threads engineering kernel (2 weeks)
+
+**Objective**: Atomic decomposition of bloat sources + addendum-thread convention + toolbox scripts + retrieval skill + Steward loop.
+
+**Sub-deliverables**:
+
+**2a — Atomic decomposition (3 days)**:
+
+- `decision-log.md` (170KB) → `docs/cards/decisions/dec-<YYYYMMDD>-<slug>.md`, one card per decision. Generated view back to `docs/living/decide/decision-log.md` via `pnpm docs:rebuild`.
+- `MEMORY.md` index lines (over 24.4KB limit) → shrink each to ≤100 chars (currently 200–400). Existing topic files in `~/.claude/memory` already atomic; mirror or symlink into `docs/cards/memory/` for cross-context discoverability.
+- `investigations.md` (61KB) → closed investigations → `docs/cards/investigations/inv-<YYYYMMDD>-<slug>.md`. Open investigations stay in `docs/ephemeral/investigations/`.
+
+**2b — SSoT discipline by doc type (~2 days)** [**SAFE-PARALLEL — can ship now, no folder-mv dependency**]
+
+Codifies §2.7 into mechanical enforcement. Differentiates by doc type instead of applying one universal "addendum thread" pattern (the original 2b proposal that the wedge-amendment incident revealed as wrong-shape — see §1 Context).
+
+**Deliverables**:
+
+- New convention doc `docs/agent-context/doc-discipline.md` (TEMP location; Play 1b moves to `docs/living/agent-context/`). Distills §2.7: doc-type table + anti-pattern table + edit-in-place mechanics + spec lifecycle states. Loaded by `agent-context-quickstart` skill as session-start read.
+
+- `scripts/check-doc-frontmatter.mjs` extension:
+  - **HARD-FAIL** on filenames matching `*-amendment-*.md`, `*-revision-*.md`, `*-update-*.md`, `*-followup-*.md` under `docs/superpowers/specs/`. Error: "Anti-pattern: edit the canonical spec in place. See `docs/agent-context/doc-discipline.md` §Anti-patterns. If genuinely intentional (rare), document rationale in decision-log + add filename to `.docs-discipline-allowlist`."
+  - **WARN** on design-spec body diffs that add a `## Amendment` heading (allowed for ADRs only; design specs are edit-in-place).
+  - **WARN** on design-spec `status: delivered` without `delivered-by:` frontmatter (proves spec-to-PR linkage).
+  - **WARN** on `## Amendment` heading in ADR body that lacks date-prefix `YYYY-MM-DD` (format check).
+  - **WARN** on `git diff` to lines older than 7 days in `docs/decision-log.md` (append-only convention).
+
+- `agent-context-quickstart` skill (already shipped Play 3a): add step 4 "Read `docs/agent-context/doc-discipline.md`" — subagents dispatched to edit any canonical doc get discipline rules on session start.
+
+- Decision-log entry convention: add `[supersedes <doc>#<section>]` marker pattern (machine-readable; greppable). Documented in doc-discipline.md.
+
+- Migration of existing scar tissue: one-shot grep for existing "PARTIALLY SUPERSEDED" / "AMENDED" prefixes in canonical specs → for each, replace prefix with proper edit-in-place + decision-log entry. (Not for ADRs — those use the existing amendment-block convention.)
+
+**Why this prevents the failure mode**: anyone (human or agent) attempting `2026-05-16-improve-tab-amendment-design.md` hits CI fail. They must edit the canonical spec in place. Convention + validator together encode the principle mechanically. New session reading the canonical spec sees current truth; no mental merging; no orphan-side-spec hunting.
+
+**Exit criteria**:
+
+- Validator extension shipped + tested against current corpus (zero false positives on existing files)
+- `doc-discipline.md` exists; `agent-context-quickstart` skill loads it
+- Manual spot-check: try creating `*-amendment-*.md` in a feature branch — CI fails as expected
+- One representative existing canonical spec gets `delivered-by:` field added (proves the convention against `status: delivered` warning)
+
+**Why now (safe-parallel)**: this is convention + validator only. No file moves, no decompositions, no shared-aggregate edits. Zero conflict risk with active wedge engineering. Ship before Plays 2a/c/d/e (which are blocked on Play 1b folder restructure).
+
+**2c — Toolbox scripts (3 days)**:
+
+- `scripts/docs-toolbox/find.mjs` — ripgrep + frontmatter filter, returns top-5 cards with excerpts.
+- `scripts/docs-toolbox/get.mjs` — full card by id.
+- `scripts/docs-toolbox/related.mjs` — graph traversal via `related:` + body `[[name]]`.
+- `scripts/docs-toolbox/recent.mjs` — new cards + amendment entries since date.
+- `scripts/docs-toolbox/verify.mjs` — bump `last-verified` + `verified-against-commit`.
+- `scripts/docs-toolbox/amend.mjs` — append to `## Amendments` section.
+- Wired to `pnpm docs:find/get/related/recent/verify/amend` (root `package.json`).
+
+**2d — Docs toolbox skill (1 day)**:
+
+- `.claude/skills/docs-toolbox/SKILL.md` with `description` triggering on doc-retrieval tasks.
+- Body: when-to-use, tool surface (`pnpm docs:*`), task kits, examples.
+- `docs/llms.txt` converts from static catalog → router pointing to this skill.
+
+**2e — Steward loop (2 days)**:
+
+- `scripts/docs-steward.mjs` — Node CLI diffs `verified-against-commit` vs current HEAD; outputs markdown report.
+- `.claude/commands/docs-steward.md` slash command + Claude triage session.
+
+### Play 3 — Skills inventory build-out (3 days + ongoing)
+
+**Objective**: Expand from 2 skills to ~18 task-typed skills covering common subagent dispatches. Closes the **4%-of-living agent-context gap**.
+
+**Tier 1 — Agent-context essentials** (close the 4% gap first):
+
+- `agent-context-quickstart` — 5-minute onboard surface
+- `package-router` — which CLAUDE.md to load for a given work area
+- `store-state-glossary` — what's in each Zustand store
+
+**Tier 2 — Task-typed procedure skills** (15 skills):
+ship-ui-primitive, add-chart, migrate-store, add-stats-test, add-i18n-string, write-spec, amend-adr, run-coscout-eval, debug-canvas-issue, add-investigation-card, add-decision-card, verify-spec-against-impl, promote-plan-to-transcript, run-docs-steward, architecture-test-write.
+
+### Play 4 — One canonical artifact per concept (1 week)
+
+**Objective**: Execute the 8 consolidation moves from Agent 2 + archive fossils + lock canonical homes.
+
+Agent 2's 8 prioritized consolidation moves:
+
+| #   | Move                                                                                                       | Effort | Risk   |
+| --- | ---------------------------------------------------------------------------------------------------------- | ------ | ------ |
+| 1   | Collapse decision-log Wedge entry to 5-sentence + cite ADR-082                                             | 1h     | Low    |
+| 2   | Compress OVERVIEW.md §The journey model 300→2 sentences + cross-link USER-JOURNEYS                         | 30m    | Low    |
+| 3   | Clarify methodology.md frontmatter on dual-source partnership                                              | 30m    | Low    |
+| 4   | Archive tier-philosophy.md + redirect banner                                                               | 1h     | Low    |
+| 5   | (negative finding) DO NOT move positioning §2.2 ("What VariScout Is Not") — keep there, link from OVERVIEW | —      | —      |
+| 6   | Add cross-links positioning.md/CLAUDE.md → business-bible H1–H6                                            | 30m    | Low    |
+| 7   | Move decision-log "Projects tab" + "8f canvas" amendments → investigations.md `[RESOLVED]`                 | 2h     | Medium |
+| 8   | Add "Canonical artifacts" section to llms.txt                                                              | 30m    | Low    |
+
+**USER-JOURNEYS variants**: Keep all 6 (audit-confirmed: mode-specific splits, not drift). Add `inherits-from: USER-JOURNEYS.md` to the 5 mode-specific variants.
+
+### Play 5 — Telemetry + 90-day archive cohort (2 days)
+
+Toolbox scripts log every query → `docs/.telemetry/queries.jsonl` (gitignored). `pnpm docs:cohort-report` surfaces untouched-in-90-days cards, top-queried, never-returned.
+
+### Play 6 — Constrain consolidation + invariants rename (2 days)
+
+- `git mv .claude/rules/ .claude/invariants/` — clarifies purpose
+- Create `.claude/INVARIANTS.md` synthesizing index
+- Per-invariant canonical home + enforcement mechanism inventory
+
+### Play 7 — System auto-generation (3 days)
+
+`pnpm docs:gen-arch` emits dependency graph (mermaid) + sub-path export map + Tailwind v4 source map from live package.json + tsconfig. Pre-push hook regenerates on each push.
+
+### Play 8 — CoScout AX-design consolidation (2 days)
+
+New canonical `docs/living/design/coscout-ax-design.md` with `topic: [ax, coscout]` gathering ADR-060/068/069 + coscout-prompts rule + eval discipline. All constituents cite back to it.
+
+---
+
+## 4. Migration Sequence
+
+> **Status as of 2026-05-17**: Phase A + Plays 1a, 3a, 6, 7, 8 SHIPPED via [PR #184](https://github.com/jukka-matti/variscout/pull/184) (13 commits, all safe-parallel with active wedge V1 engineering). Play 1 was split into 1a (foundation + schema collapse, shipped) + 1b (521-doc folder restructure, deferred). Play 2 was split into 2b (SSoT discipline, **now safe-parallel — ship next**) + 2a/2c/2d/2e (cards + toolbox + steward, deferred). Plays 1b, 2a/c/d/e, 4 await post-wedge-V1 quiet window. The revised 2b is the lesson from the wedge-amendment incident baked back into the strategy — see §1 Context.
+
+### Ideal sequence (with current split + deferrals annotated)
+
+| Week  | Phase / Plays                                         | Effort     | Status                      | Output                                                                                                               |
+| ----- | ----------------------------------------------------- | ---------- | --------------------------- | -------------------------------------------------------------------------------------------------------------------- |
+| **0** | **Phase A quick wins**                                | ~3 hours   | ✅ SHIPPED                  | Wedge content consolidated; zero-infra warm-up                                                                       |
+| 1     | Play 1a (foundation artifacts + schema collapse)      | 1 day      | ✅ SHIPPED                  | Memo + spec + ADR-083; 22→4 STATUS; alias maps; frontmatter backfill                                                 |
+| 1     | Play 1b (521-doc folder restructure)                  | 2 days     | ⏸ DEFERRED                  | `git mv` to `docs/stable/` + `docs/living/<purpose>/` + `docs/cards/`. Wait for quiet wedge window.                  |
+| 1     | **Play 2b (SSoT discipline by doc type)**             | ~2 days    | 🟡 **NEXT (safe-parallel)** | Validator forbidding `*-amendment-*.md`; `doc-discipline.md` convention; lifecycle state enforcement. **Ship now.**  |
+| 1–3   | Play 2 (2a + 2c + 2d + 2e: cards + toolbox + steward) | ~1.5 weeks | ⏸ DEFERRED                  | Blocked on Play 1b substrate + quiet repo (decomposing decision-log while wedge amends it = constant rework).        |
+| 3     | Play 3a (Tier 1 agent-context skills)                 | 1 day      | ✅ SHIPPED                  | `agent-context-quickstart`, `package-router`, `store-state-glossary` — closes 4% agent-context gap                   |
+| 3     | Play 3b (Tier 2 procedure skills, ~15 skills)         | 3 days+    | ⏸ later session             | `ship-ui-primitive`, `add-chart`, `migrate-store`, etc. Build as dispatch logs justify (data-driven prioritization). |
+| 3     | Play 6 (invariants rename + INVARIANTS.md)            | 1 day      | ✅ SHIPPED                  | `.claude/rules/` → `.claude/invariants/`; 15 hard + 10 soft + 6 topic-scoped synthesized                             |
+| 4     | Play 4 (one-canonical consolidation)                  | 1 week     | ⏸ DEFERRED                  | Agent 2's remaining 4 of 8 moves. Blocked on Play 1b + Play 2.                                                       |
+| 5     | Play 5 (telemetry + 90-day cohort report)             | 2 days     | ⏸ blocked on 2c             | Toolbox query logging → cohort report. Useful after 2c lands.                                                        |
+| 5     | Play 7 (`pnpm docs:gen-arch`)                         | 1 day      | ✅ SHIPPED                  | Dep graph + sub-path export map + Tailwind @source coverage; pre-push staleness check                                |
+| 5     | Play 8 (CoScout AX-design consolidation)              | 1 day      | ✅ SHIPPED                  | Canonical `coscout-ax-design.md`; cross-links from ADR-060/068/069 + prompt invariant                                |
+
+### Lessons baked back into the strategy (2026-05-17)
+
+- **Original Play 2b ("addendum threads on living docs") was wrong-shape** — discovered when a wedge-spec amendment was captured as a separate `*-amendment-design.md` file, invisible to a new session reading only the canonical wedge spec. Revised 2b differentiates by doc type (§2.7).
+- **Mid-session TaskStop pivot worked cleanly** — when wedge-conflict surfaced, killed in-flight Play 1 subagent + `git reset --hard` to safe commit; split into safe (1a) + deferred (1b). See `feedback_taskstop_subagent_pivot`.
+- **Auto-generated docs need `.prettierignore`** — pre-push staleness check vs lint-staged prettier conflict during Play 7. See `feedback_autogen_doc_prettierignore`.
+- **Doc-strategy work itself was the testbed** — Play 1 subagent dispatched without checking active parallel writers (wedge V1 engineering); user surfaced; corrected. See `feedback_parallel_workstream_conflict_check`.
+
+---
+
+## 5. Critical Files
+
+### To create (Play 1)
+
+- `docs/stable/`, `docs/living/<purpose>/`, `docs/ephemeral/`, `docs/cards/` hierarchy
+- This spec (`2026-05-16-docs-strategy-design.md`)
+- `docs/superpowers/specs/2026-05-16-docs-strategy-memo.md`
+- `docs/07-decisions/adr-083-eight-purpose-doc-taxonomy.md`
+
+### To modify (Play 1)
+
+- `scripts/docs-frontmatter-schema.mjs` — collapse schema
+- `scripts/check-doc-frontmatter.mjs` — enforce new fields + alias maps
+- `scripts/docs-frontmatter-fix.mjs` — backfill `purpose` + `tier`
+- `docs/llms.txt` — update paths
+- Root `CLAUDE.md` — "Where to look" section
+
+### To archive (confirmed by Agent 1 audit)
+
+- `docs/09-baseline/` → `docs/archive/baselines/` — pre-wedge audit snapshots
+- `docs/10-development/project-status-audit-2026-04-16.md` → `docs/archive/audits/`
+- `docs/08-products/tier-philosophy.md` → `docs/archive/` (Play 4c)
+- `.claude/agents/flow-nexus/` → `.claude/agents/experimental/` (verify inbound refs first)
+
+### To delete (Play 4g)
+
+- `docs/03-features/learning/glossary.md` — merge unique content into root `glossary.md`, then `git rm`
+
+---
+
+## 6. Verification
+
+**Play 1 exit criteria**:
+
+- `pnpm docs:health` green with new schema
+- All inter-doc links resolve (no broken refs)
+- `bash scripts/pr-ready-check.sh` green
+- `git log --follow` works on moved files
+
+**Overall health indicator** (30 days post-Play-5):
+
+- Cohort report shows non-zero queries logged
+- ≥1 Steward run completed
+- `last-verified` rate on `living` tier >50%
+
+---
+
+## 7. Audit Findings Summary
+
+### 7.1 Purpose distribution (Agent 1 findings)
+
+| Purpose       | Stable | Living | Ephemeral | Total                      |
+| ------------- | ------ | ------ | --------- | -------------------------- |
+| orient        | 12     | 0      | 0         | 12                         |
+| decide        | 0      | 90     | 0         | 90                         |
+| design        | 0      | 180    | 5         | 185                        |
+| build         | 0      | 50     | 2         | 52                         |
+| system        | 0      | 20     | 5         | 25                         |
+| constrain     | 8      | 0      | 0         | 8                          |
+| agent-context | 0      | **15** | 0         | **15** (4% — critical gap) |
+| remember      | 100    | 0      | 0         | 100                        |
+
+### 7.2 Multi-canonical overlap matrix (Agent 2 findings)
+
+Highest-overlap pair: decision-log "Wedge pivot" entry (~2000 words) vs ADR-082 — 90% verbatim duplication. Canonical home: **ADR-082**.
+
+**USER-JOURNEYS variants verdict: KEEP ALL SIX** — mode-specific methodology, not stale drift.
+
+### 7.3 Implications resolved
+
+- Investigations.md disposition: keeps identity with OPEN (decide) vs [RESOLVED] (remember) split. Closed → cards via Play 2a.
+- Methodology canonical: dual (vision spec §2.1 = spine, methodology.md = pedagogical narrative).
+- Fossils: `09-baseline/`, `project-status-audit-2026-04-16.md`, `03-features/learning/glossary.md`, `tier-philosophy.md` confirmed.
+
+---
+
+## 10. Risks & Mitigations
+
+| Risk                                                                             | Mitigation                                                                                                                                                                                               |
+| -------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| Mass `git mv` breaks link resolution across 521 docs                             | `scripts/check-dead-links.sh` runs as gate; staged batches per purpose; fix broken links per batch                                                                                                       |
+| Frontmatter migration loses metadata                                             | `docs-frontmatter-fix.mjs` runs dry-first; diff inspected; backup branch before merge                                                                                                                    |
+| Atomic decomposition of `decision-log.md` loses context                          | Generated view preserves narrative; manual review of 5 sample decisions post-decomposition                                                                                                               |
+| Steward loop produces noisy false-positives                                      | Conservative first cut (only flag cards where cited symbols are gone); tune threshold post-Play-5                                                                                                        |
+| 15 skills is too many to maintain                                                | Start with 8 highest-frequency tasks; add others as dispatch logs justify; `pnpm docs:cohort-report` tracks unused skills                                                                                |
+| AX-design doc for CoScout duplicates ADR content                                 | Strict citation discipline: AX-design doc states current state + tradeoffs; ADRs remain decision provenance                                                                                              |
+| Schema collapse breaks downstream tooling assuming 22 STATUS values              | `docs-frontmatter-fix.mjs` maps old → new; transitional alias period (old values warn but pass) for 1 cycle                                                                                              |
+| Doc-discipline drift if not mechanically enforced (the wedge-amendment incident) | Play 2b validator HARD-FAILs `*-amendment-*.md` filenames; WARNs on body-diff anti-patterns; `agent-context-quickstart` loads `doc-discipline.md` on session start so every subagent gets the rules cold |
+| Validator false positives block legitimate work                                  | `.docs-discipline-allowlist` escape hatch for rare intentional exceptions; documented in `doc-discipline.md`; allowlist additions require decision-log entry citing the rationale                        |
+| Spec body becomes stale during long active-status periods                        | `last-verified` frontmatter + Play 5 cohort report flags `living`-tier specs unverified >90 days; Steward loop (Play 2e) proposes amendment thread or `delivered` transition                             |
diff --git a/docs/superpowers/specs/2026-05-16-docs-strategy-memo.md b/docs/superpowers/specs/2026-05-16-docs-strategy-memo.md
new file mode 100644
index 000000000..68fbbf4df
--- /dev/null
+++ b/docs/superpowers/specs/2026-05-16-docs-strategy-memo.md
@@ -0,0 +1,86 @@
+---
+title: 'VariScout Docs Strategy 2026 — Memo'
+status: active
+date: 2026-05-16
+last-verified: 2026-05-17
+purpose: orient
+tier: stable
+audience: both
+topic: [docs-strategy, meta]
+related: [2026-05-16-docs-strategy-design, adr-083-eight-purpose-doc-taxonomy]
+---
+
+# VariScout Docs Strategy 2026 — Memo
+
+_CTO summary. Full design: [2026-05-16-docs-strategy-design.md](2026-05-16-docs-strategy-design.md). Schema change: [ADR-083](../../07-decisions/adr-083-eight-purpose-doc-taxonomy.md)._
+
+---
+
+## What we're changing
+
+VariScout's docs scale-friction problem is **over-discipline at the metadata layer, under-discipline at the lifecycle layer**. We're flipping that: collapse the metadata schema, decompose the actual bloat sources, expose retrieval as a skill, replace batch audits with a steward loop, expand the AX-dev surface (skills) to 2026 conventions, and **codify single-source-of-truth discipline by doc type** (full design §2.7) — design specs edit-in-place, ADRs amendment-block-at-bottom, decision-log append-only — mechanically enforced via validator (no `*-amendment-*.md` orphan files).
+
+## Diagnosis (from the brainstorm audit)
+
+- 487 active docs across 521 files (94% utilization)
+- Distribution: **Design 51%**, Decide 25%, Build 14%, System 6%, **Agent-context 4% — the critical gap**
+- 17 emergent topic tags (close to my proposed vocabulary; codified in the new schema)
+- 6 USER-JOURNEYS variants confirmed as **legitimate mode-specific splits** (not stale drift) — keep all
+- Wedge content (2026-05-16) is the highest-impact consolidation target: decision-log replicates 90% of ADR-082 verbatim (~2000 redundant words)
+- Methodology coverage is fragmented but recoverable: vision spec §2.1 (spine) + methodology.md (narrative) should be paired companions, not competing canonicals
+- `.claude/rules/` should rename to `.claude/invariants/` — confirmed all 6 current files are constraint-purpose
+- Specific fossils confirmed: `09-baseline/`, `10-development/project-status-audit-2026-04-16.md`, `03-features/learning/glossary.md`, `08-products/tier-philosophy.md`
+
+## The foundation
+
+Every doc serves exactly one of **8 purposes** (orient, decide, design, build, system, constrain, agent-context, remember) at one of **4 tiers** (stable, living, ephemeral, card). Topic tags (`ux`, `ax`, `coscout`, `canvas`, …) discriminate within purposes.
+
+Industry frameworks (Diátaxis, Google's engineering-practices model) are human-reader-centric and pre-LLM-era. VariScout's 8-purpose taxonomy explicitly covers both human readers AND the AX-dev surface (subagents + CoScout) that's now half its doc system.
+
+## The 8 plays + Phase A warm-up (~5.5 weeks elapsed)
+
+- **Phase A** (~3h) — Quick wins: 4 of Agent 2's 8 consolidation moves (wedge content collapse, OVERVIEW compression, methodology clarification, business-bible cross-links). Zero-infra warm-up before structural moves.
+
+1. **Velocity-tier restructure** (3d) — `git mv` to tier folders + schema collapse (22→4 STATUS, 18→tags, 14→3 AUDIENCE); fossil archive batch
+2. **Cards & Discipline** (2w total; split shipped) — Play **2b (SSoT discipline by doc type)** can ship now safe-parallel (validator forbidding `*-amendment-*.md`; `doc-discipline.md` convention; lifecycle state enforcement). Plays **2a + 2c + 2d + 2e** (atomic decomposition + `pnpm docs:*` toolbox + `docs-toolbox` skill + `/docs-steward` loop) blocked on Play 1b substrate.
+3. **Skills inventory build-out** (3d + ongoing) — expand from 2 to **18+** task-typed skills, **with Tier 1 agent-context essentials first** (`agent-context-quickstart`, `package-router`, `store-state-glossary`) to close the 4% gap immediately
+4. **One canonical per concept** (1w) — execute remaining 4 of Agent 2's 8 moves; tier-philosophy archive; USER-JOURNEYS variant `inherits-from:` lineage; mis-housed folder refactors (08-products split, 06-design-system split)
+5. **Telemetry + 90-day archive cohort** (2d) — toolbox skill logs queries; auto-prune unread cohort
+6. **Constrain consolidation + invariants rename** (2d) — `.claude/rules/` → `.claude/invariants/` + `.claude/INVARIANTS.md` synthesis; collapse 4–5-place duplication
+7. **System auto-generation** (3d) — `pnpm docs:gen-arch` from package.json + tsconfig
+8. **CoScout AX-design consolidation** (2d) — gather ADR-060/068/069 + prompt rule into single canonical `coscout-ax-design.md` with `topic: [ax, coscout]`
+
+## The 2026 differentiator
+
+**The Steward Loop** (`/docs-steward`) — periodic Claude session that diffs `verified-against-commit` against current HEAD, proposes amendments for stale cards, outputs a review-ready report. The pending **Phase C audit (~500 docs)** becomes a continuous background process, not a 500-doc cliff.
+
+## What we kill
+
+- 22 STATUS enums → 4
+- 18 CATEGORY enums → free-form topic tags
+- 14 AUDIENCE enums → 3
+- 11 numbered folders → 3 velocity tiers + cards/
+- `.claude/rules/` (rename to `.claude/invariants/` for purpose clarity)
+- ~2000 words of decision-log Wedge entry duplication (collapses to cite ADR-082)
+- ~300 words of OVERVIEW.md methodology recap (compresses to 2 sentences + cross-link)
+- Orphan `*-amendment-design.md` side-spec files → edit canonical in place + decision-log entry (validator HARD-FAILs filename anti-pattern)
+- "⚠️ PARTIALLY SUPERSEDED" prefixes on canonical specs → spec body edited in place + supersession captured in decision-log + frontmatter
+- `llms.txt` as static catalog → routing skill (Play 2d)
+- Phase C audit as future project → Steward as weekly habit
+- Specific fossils: `09-baseline/`, `tier-philosophy.md`, `10-development/project-status-audit-2026-04-16.md`, `03-features/learning/glossary.md`
+
+## What we preserve (audit-confirmed)
+
+- ADR discipline (74 of them is healthy; convention works)
+- All 6 USER-JOURNEYS variants (legitimate mode-specific splits)
+- Memory atomic-notes pattern (promoted up to docs)
+- Validation tooling (extended, not replaced)
+- Frontmatter SSOT in `scripts/` (pattern is right; schema simplifies)
+- Subagent-driven-development workflow (materially improved by Cards & Threads + Skills inventory)
+- Nested CLAUDE.md per package (progressive disclosure is correct)
+- decision-log as the re-litigation gate (decompose storage, keep function)
+- methodology.md + vision spec §2.1 dual-source partnership (clarified, not collapsed)
+
+## CTO verdict
+
+Current strategy is sound but is **scaling its own friction**: the more docs, the more enums to maintain, the more aggregates to keep current, the more "partially superseded" scar tissue. The fix is to invert that — make adding-a-doc _cheaper_ (smaller schema, atomic cards) and make doc-rot _self-detecting_ (Steward + telemetry). The agent-context cohort is the highest-leverage cohort to expand (4% of living docs, read every dispatch). Investment ~5.5 weeks (Phase A + 8 plays); recurring authoring tax drops materially; Phase C cliff is dissolved into a continuous loop; ~3 hours of Phase A warm-up removes the most visible noise immediately.
diff --git a/docs/superpowers/specs/2026-05-16-wedge-architecture-design.md b/docs/superpowers/specs/2026-05-16-wedge-architecture-design.md
index d86b25476..619020a08 100644
--- a/docs/superpowers/specs/2026-05-16-wedge-architecture-design.md
+++ b/docs/superpowers/specs/2026-05-16-wedge-architecture-design.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: Wedge architecture — single-product VariScout for improvement specialists
-audience: [product, engineer, designer, business]
+audience: human
 category: design-spec
 status: draft
 last-reviewed: 2026-05-16
diff --git a/docs/superpowers/specs/index.md b/docs/superpowers/specs/index.md
index 09ff993da..a89504a1b 100644
--- a/docs/superpowers/specs/index.md
+++ b/docs/superpowers/specs/index.md
@@ -1,6 +1,8 @@
 ---
+tier: living
+purpose: design
 title: 'Design Specs (superpowers)'
-status: reference
+status: active
 ---
 
 # Design Specs (superpowers)
diff --git a/package.json b/package.json
index 857b43776..4fd2610e6 100644
--- a/package.json
+++ b/package.json
@@ -23,7 +23,8 @@
     "docs:sync": "bash scripts/sync-monorepo-counts.sh",
     "codex:ruflo-check": "bash scripts/check-codex-ruflo.sh",
     "check:boundaries": "bash scripts/check-level-boundaries.sh",
-    "check:i18n": "npx tsx scripts/check-i18n.ts"
+    "check:i18n": "npx tsx scripts/check-i18n.ts",
+    "docs:gen-arch": "node scripts/docs/gen-arch.mjs"
   },
   "lint-staged": {
     "**/*.{ts,tsx}": [
diff --git a/scripts/check-doc-frontmatter.mjs b/scripts/check-doc-frontmatter.mjs
index 858107bf6..b38775269 100755
--- a/scripts/check-doc-frontmatter.mjs
+++ b/scripts/check-doc-frontmatter.mjs
@@ -1,19 +1,28 @@
 #!/usr/bin/env node
 // check-doc-frontmatter.mjs — validate docs/** frontmatter against the schema.
-// Warn mode until CUTOFF_DATE; after that, exit 1 on violations.
-// Mirrors the pattern in scripts/check-dead-links.sh.
+// ADR-083 (2026-05-16): transitional alias maps for STATUS + AUDIENCE.
+// Old STATUS/AUDIENCE values warn; unknown PURPOSE/TIER hard-fail when present.
+// --report mode includes count of docs missing purpose + tier fields.
 
 import { readFileSync, readdirSync, statSync } from 'node:fs';
 import { join, relative } from 'node:path';
 import { fileURLToPath } from 'node:url';
 import { dirname } from 'node:path';
 import { parse as parseYaml } from 'yaml';
-import { schema, classify } from './docs-frontmatter-schema.mjs';
+import {
+  schema,
+  classify,
+  STATUS_ALIAS_MAP,
+  AUDIENCE_ALIAS_MAP,
+  PURPOSE,
+  TIER,
+} from './docs-frontmatter-schema.mjs';
 
 const ROOT = dirname(dirname(fileURLToPath(import.meta.url)));
 const DOCS = join(ROOT, 'docs');
 const CUTOFF_DATE = process.env.CUTOFF_DATE ?? '2026-05-15';
 const TODAY = new Date().toISOString().slice(0, 10);
+const REPORT_MODE = process.argv.includes('--report');
 
 function walk(dir, out = []) {
   for (const entry of readdirSync(dir)) {
@@ -42,6 +51,12 @@ const violations = {
   unknownEnum: [],
   casingDrift: [],
   malformedYaml: [],
+  // Warnings (old-value aliases) — don't fail the build
+  aliasedStatus: [],
+  aliasedAudience: [],
+  // Report-only stats
+  missingPurpose: [],
+  missingTier: [],
 };
 
 function check(file) {
@@ -64,64 +79,131 @@ function check(file) {
     return;
   }
 
+  // Required fields
   for (const key of rules.required) {
     if (fm[key] == null || fm[key] === '') {
       violations.missingRequired.push(`${rel}: missing '${key}' [${kind}]`);
     }
   }
 
+  // Enum validation with alias-map awareness
   for (const [field, allowed] of Object.entries(rules.enums ?? {})) {
     const values = asArray(fm[field]);
     for (const v of values) {
       if (v == null || v === '') continue;
       const str = String(v);
+
       if (allowed.includes(str)) continue;
+
       const lower = str.toLowerCase();
+
+      // Check casing drift against canonical values
       if (allowed.includes(lower)) {
         violations.casingDrift.push(`${rel}: ${field}='${str}' should be '${lower}'`);
-      } else {
-        violations.unknownEnum.push(`${rel}: ${field}='${str}' not in allowed set`);
+        continue;
+      }
+
+      // Check alias maps for STATUS and AUDIENCE (warn, not fail)
+      if (field === 'status' && STATUS_ALIAS_MAP[lower] != null) {
+        violations.aliasedStatus.push(
+          `${rel}: status='${str}' → aliased to '${STATUS_ALIAS_MAP[lower]}' (transitional)`,
+        );
+        continue;
+      }
+      if (field === 'audience' && AUDIENCE_ALIAS_MAP[lower] != null) {
+        violations.aliasedAudience.push(
+          `${rel}: audience='${str}' → aliased to '${AUDIENCE_ALIAS_MAP[lower]}' (transitional)`,
+        );
+        continue;
       }
+
+      // PURPOSE and TIER hard-fail when present but unknown
+      if (field === 'purpose' && !PURPOSE.includes(str)) {
+        violations.unknownEnum.push(`${rel}: purpose='${str}' not in allowed set ${PURPOSE.join('|')}`);
+        continue;
+      }
+      if (field === 'tier' && !TIER.includes(str)) {
+        violations.unknownEnum.push(`${rel}: tier='${str}' not in allowed set ${TIER.join('|')}`);
+        continue;
+      }
+
+      // CATEGORY is no longer in the schema; skip silently (old docs may have it)
+      if (field === 'category') continue;
+
+      violations.unknownEnum.push(`${rel}: ${field}='${str}' not in allowed set`);
     }
   }
+
+  // Report-only: track missing purpose + tier
+  if (fm['purpose'] == null || fm['purpose'] === '') {
+    violations.missingPurpose.push(rel);
+  }
+  if (fm['tier'] == null || fm['tier'] === '') {
+    violations.missingTier.push(rel);
+  }
 }
 
 const files = walk(DOCS);
 for (const f of files) check(f);
 
-const total =
+const hardViolationTotal =
   violations.missingFrontmatter.length +
   violations.missingRequired.length +
   violations.unknownEnum.length +
   violations.casingDrift.length +
   violations.malformedYaml.length;
 
+const warningTotal = violations.aliasedStatus.length + violations.aliasedAudience.length;
+
 function report() {
-  if (total === 0) {
+  if (REPORT_MODE) {
+    console.log(`\n=== Frontmatter Report: ${files.length} docs ===`);
+    console.log(`  Missing purpose: ${violations.missingPurpose.length} docs`);
+    console.log(`  Missing tier:    ${violations.missingTier.length} docs`);
+    console.log(`  Aliased status:  ${violations.aliasedStatus.length} (transitional; run docs-frontmatter-fix to resolve)`);
+    console.log(`  Aliased audience:${violations.aliasedAudience.length} (transitional; run docs-frontmatter-fix to resolve)`);
+    console.log(`  Hard violations: ${hardViolationTotal}`);
+    console.log('');
+  }
+
+  if (hardViolationTotal === 0 && warningTotal === 0) {
     console.log(`✓ Frontmatter check: ${files.length} docs validated, no issues.`);
     return;
   }
 
-  const show = (label, list, limit = 10) => {
+  const show = (label, list, limit = 10, isWarn = false) => {
     if (list.length === 0) return;
-    console.error(`\n${label} (${list.length}):`);
+    const prefix = isWarn ? '⚠' : '✗';
+    console.error(`\n${prefix} ${label} (${list.length}):`);
     for (const item of list.slice(0, limit)) console.error(`  ${item}`);
     if (list.length > limit) console.error(`  … ${list.length - limit} more`);
   };
 
-  console.error(`⚠ Frontmatter check: ${total} issue(s) across ${files.length} docs.`);
-  show('✗ Missing frontmatter', violations.missingFrontmatter);
-  show('✗ Missing required field', violations.missingRequired);
-  show('✗ Unknown enum value', violations.unknownEnum);
-  show('✗ Casing drift (auto-fixable)', violations.casingDrift);
-  show('✗ Malformed YAML', violations.malformedYaml);
-  console.error(
-    `\nSchema: scripts/docs-frontmatter-schema.mjs. ` +
-      `After ${CUTOFF_DATE}, this check fails the commit.`,
-  );
+  if (hardViolationTotal > 0) {
+    console.error(`⚠ Frontmatter check: ${hardViolationTotal} hard issue(s) + ${warningTotal} alias warning(s) across ${files.length} docs.`);
+    show('Missing frontmatter', violations.missingFrontmatter);
+    show('Missing required field', violations.missingRequired);
+    show('Unknown enum value (hard-fail)', violations.unknownEnum);
+    show('Casing drift (auto-fixable)', violations.casingDrift);
+    show('Malformed YAML', violations.malformedYaml);
+  }
+
+  show('Aliased status (transitional — run docs-frontmatter-fix.mjs)', violations.aliasedStatus, 5, true);
+  show('Aliased audience (transitional — run docs-frontmatter-fix.mjs)', violations.aliasedAudience, 5, true);
+
+  if (hardViolationTotal > 0) {
+    console.error(
+      `\nSchema: scripts/docs-frontmatter-schema.mjs. ` +
+        `After ${CUTOFF_DATE}, this check fails the commit.`,
+    );
+  }
+
+  if (warningTotal > 0 && hardViolationTotal === 0) {
+    console.log(`✓ Frontmatter check: ${files.length} docs validated, ${warningTotal} transitional alias warning(s). Run docs-frontmatter-fix.mjs to resolve.`);
+  }
 }
 
 report();
 
-if (TODAY >= CUTOFF_DATE && total > 0) process.exit(1);
+if (TODAY >= CUTOFF_DATE && hardViolationTotal > 0) process.exit(1);
 process.exit(0);
diff --git a/scripts/docs-frontmatter-fix.mjs b/scripts/docs-frontmatter-fix.mjs
index 122ef3edd..b875236c3 100644
--- a/scripts/docs-frontmatter-fix.mjs
+++ b/scripts/docs-frontmatter-fix.mjs
@@ -1,13 +1,18 @@
 #!/usr/bin/env node
 // docs-frontmatter-fix.mjs — one-shot normalization codemod.
-// Safe, narrow fixes only:
+// ADR-083 (2026-05-16): extended with purpose + tier backfill + alias map resolution.
+//
+// Fixes applied:
 //   1. Lowercase STATUS enum values ('Accepted' → 'accepted').
-//   2. Resolve the 'Superseded by ADR-037' anomaly in adr-024.
-//   3. Add a minimal frontmatter block to the 3 bare plan files.
+//   2. Map old STATUS alias values to new canonical (accepted → active, etc.).
+//   3. Map old AUDIENCE alias values to new canonical (developer → human, etc.).
+//   4. Resolve the 'Superseded by ADR-037' anomaly in adr-024.
+//   5. Add a minimal frontmatter block to the 3 bare plan files.
+//   6. Backfill `purpose` + `tier` based on path heuristics (new in ADR-083).
+//   7. Add `archived-on` + `archived-reason` to docs in docs/archive/.
 //
 // NOT done here: filling missing required fields on general docs
-// (audience/category/status). Those require human judgment — run
-// the validator and address the `missing-required` list manually.
+// (audience/status) that require human judgment.
 //
 // Usage: node scripts/docs-frontmatter-fix.mjs [--apply]
 //   --apply writes changes. Default is dry-run.
@@ -16,7 +21,7 @@ import { readFileSync, writeFileSync, readdirSync, statSync } from 'node:fs';
 import { join, relative } from 'node:path';
 import { fileURLToPath } from 'node:url';
 import { dirname } from 'node:path';
-import { STATUS } from './docs-frontmatter-schema.mjs';
+import { STATUS, STATUS_ALIAS_MAP, AUDIENCE_ALIAS_MAP } from './docs-frontmatter-schema.mjs';
 
 const ROOT = dirname(dirname(fileURLToPath(import.meta.url)));
 const DOCS = join(ROOT, 'docs');
@@ -37,7 +42,76 @@ function change(file, desc) {
   changes.push(`  ${relative(ROOT, file)}: ${desc}`);
 }
 
-// Fix 1: lowercase status values
+// === Purpose + tier heuristics based on path ===
+// Returns { purpose, tier, topicHint } or null if no heuristic matches.
+function inferPurposeTier(relPath) {
+  // New structure (post-Play-1 moves)
+  if (/docs\/living\/decide\/adrs\//.test(relPath)) return { purpose: 'decide', tier: 'living' };
+  if (/docs\/living\/decide\//.test(relPath)) return { purpose: 'decide', tier: 'living' };
+  if (/docs\/living\/design\/specs\//.test(relPath)) return { purpose: 'design', tier: 'living' };
+  if (/docs\/living\/design\/journeys\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,journey' };
+  if (/docs\/living\/design\/features\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,feature' };
+  if (/docs\/living\/design\/cases\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'use-case' };
+  if (/docs\/living\/design\/design-system\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,design-system' };
+  if (/docs\/living\/design\/tutorials\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,tutorial' };
+  if (/docs\/living\/design\//.test(relPath)) return { purpose: 'design', tier: 'living' };
+  if (/docs\/living\/system\//.test(relPath)) return { purpose: 'system', tier: 'living' };
+  if (/docs\/living\/build\//.test(relPath)) return { purpose: 'build', tier: 'living' };
+  if (/docs\/living\/orient\//.test(relPath)) return { purpose: 'orient', tier: 'living' };
+  if (/docs\/stable\/vision\//.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/stable\/products\//.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/stable\//.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/ephemeral\/plans\//.test(relPath)) return { purpose: 'build', tier: 'ephemeral' };
+  if (/docs\/ephemeral\/investigations\//.test(relPath))
+    return { purpose: 'decide', tier: 'ephemeral' };
+  if (/docs\/ephemeral\/transcripts\//.test(relPath))
+    return { purpose: 'remember', tier: 'ephemeral' };
+  if (/docs\/cards\/decisions\//.test(relPath)) return { purpose: 'decide', tier: 'card' };
+  if (/docs\/cards\/memory\//.test(relPath)) return { purpose: 'remember', tier: 'card' };
+  if (/docs\/cards\/investigations\//.test(relPath)) return { purpose: 'decide', tier: 'card' };
+  if (/docs\/archive\//.test(relPath)) return { purpose: 'remember', tier: 'stable' };
+
+  // Legacy paths (pre-Play-1; still present during transition)
+  if (/docs\/07-decisions\/adr-/.test(relPath)) return { purpose: 'decide', tier: 'living' };
+  if (/docs\/07-decisions\//.test(relPath)) return { purpose: 'decide', tier: 'living' };
+  if (/docs\/superpowers\/specs\//.test(relPath)) return { purpose: 'design', tier: 'living' };
+  if (/docs\/superpowers\/plans\//.test(relPath)) return { purpose: 'build', tier: 'ephemeral' };
+  if (/docs\/01-vision\//.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/02-journeys\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,journey' };
+  if (/docs\/03-features\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,feature' };
+  if (/docs\/04-cases\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'use-case' };
+  if (/docs\/05-technical\//.test(relPath)) return { purpose: 'system', tier: 'living' };
+  if (/docs\/06-design-system\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,design-system' };
+  if (/docs\/08-products\//.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/09-baseline\//.test(relPath)) return { purpose: 'remember', tier: 'stable' };
+  if (/docs\/09-tutorials\//.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,tutorial' };
+  if (/docs\/10-development\//.test(relPath)) return { purpose: 'build', tier: 'living' };
+
+  // Root-level special docs
+  if (/docs\/decision-log\.md$/.test(relPath)) return { purpose: 'decide', tier: 'living' };
+  if (/docs\/investigations\.md$/.test(relPath)) return { purpose: 'decide', tier: 'ephemeral' };
+  if (/docs\/glossary\.md$/.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/OVERVIEW\.md$/.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/index\.md$/.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/roadmap\.md$/.test(relPath)) return { purpose: 'orient', tier: 'stable' };
+  if (/docs\/DATA-FLOW\.md$/.test(relPath)) return { purpose: 'system', tier: 'living' };
+  if (/docs\/USER-JOURNEYS/.test(relPath))
+    return { purpose: 'design', tier: 'living', topicHint: 'ux,journey' };
+
+  return null;
+}
+
+// Fix 1: lowercase status values (canonical set only)
 function fixStatusCasing(file, src) {
   const m = src.match(/^(---\r?\n)([\s\S]*?)(\r?\n---\r?\n)/);
   if (!m) return src;
@@ -50,16 +124,77 @@ function fixStatusCasing(file, src) {
       if (val === lower) return full;
       // Skip the ADR-024 anomaly — handled separately below.
       if (val.startsWith('Superseded by')) return full;
-      if (!STATUS.includes(lower)) return full;
+      if (!STATUS.includes(lower) && !STATUS_ALIAS_MAP[lower]) return full;
       changed = true;
-      change(file, `status '${val}' → '${lower}'`);
+      change(file, `status casing '${val}' → '${lower}'`);
       return `${prefix}${quote}${lower}${quote}`;
     },
   );
   return changed ? open + newBody + close + src.slice(m[0].length) : src;
 }
 
-// Fix 2: adr-024 anomaly — status value is a sentence
+// Fix 2: resolve STATUS alias values (accepted → active, etc.)
+function fixStatusAlias(file, src) {
+  const m = src.match(/^(---\r?\n)([\s\S]*?)(\r?\n---\r?\n)/);
+  if (!m) return src;
+  const [, open, body, close] = m;
+  let changed = false;
+  const newBody = body.replace(
+    /^(status:\s*)(['"]?)([a-z][\w-]*)\2\s*$/gm,
+    (full, prefix, quote, val) => {
+      const mapped = STATUS_ALIAS_MAP[val];
+      if (!mapped) return full;
+      changed = true;
+      change(file, `status alias '${val}' → '${mapped}'`);
+      return `${prefix}${quote}${mapped}${quote}`;
+    },
+  );
+  return changed ? open + newBody + close + src.slice(m[0].length) : src;
+}
+
+// Fix 3: resolve AUDIENCE alias values (developer → human, etc.)
+// Handles both scalar (audience: engineer) and array (audience: [analyst, engineer]) forms.
+function fixAudienceAlias(file, src) {
+  const m = src.match(/^(---\r?\n)([\s\S]*?)(\r?\n---\r?\n)/);
+  if (!m) return src;
+  const [, open, body, close] = m;
+  let changed = false;
+
+  // Scalar form: audience: engineer
+  let newBody = body.replace(
+    /^(audience:\s*)(['"]?)([a-z][\w-]*)\2\s*$/gm,
+    (full, prefix, quote, val) => {
+      const mapped = AUDIENCE_ALIAS_MAP[val];
+      if (!mapped) return full;
+      changed = true;
+      change(file, `audience alias '${val}' → '${mapped}'`);
+      return `${prefix}${quote}${mapped}${quote}`;
+    },
+  );
+
+  // Inline array form: audience: [analyst, engineer] or audience: [analyst, engineer, ...]
+  // Collapse any multi-value old-enum audience to `both` (agent values would have been manually set).
+  newBody = newBody.replace(
+    /^(audience:\s*)\[([^\]]+)\]\s*$/gm,
+    (full, prefix, inner) => {
+      const vals = inner.split(',').map((s) => s.trim().replace(/['"]/g, ''));
+      const allOld = vals.every((v) => AUDIENCE_ALIAS_MAP[v] != null || v === 'human' || v === 'agent' || v === 'both');
+      if (!allOld) return full;
+      // If already all canonical values, pass through
+      const allCanonical = vals.every((v) => v === 'human' || v === 'agent' || v === 'both');
+      if (allCanonical) return full;
+      // Determine new value: if mix includes agent-facing terms, use 'both', else 'human'
+      const newVal = 'human'; // All old audience values map to human
+      changed = true;
+      change(file, `audience array [${vals.join(', ')}] → '${newVal}'`);
+      return `${prefix}${newVal}`;
+    },
+  );
+
+  return changed ? open + newBody + close + src.slice(m[0].length) : src;
+}
+
+// Fix 4: adr-024 anomaly — status value is a sentence
 function fixAdr024Anomaly(file, src) {
   if (!file.endsWith('adr-024-scouting-report.md')) return src;
   if (!/status:\s*['"]?Superseded by ADR-037['"]?/.test(src)) return src;
@@ -70,7 +205,7 @@ function fixAdr024Anomaly(file, src) {
   );
 }
 
-// Fix 3: add minimal frontmatter to the 3 bare plan files
+// Fix 5: add minimal frontmatter to the 3 bare plan files
 const BARE_PLAN_DEFAULTS = {
   'agent-docs-architecture-phase1-foundation':
     'Agent Docs Architecture — Phase 1: Foundation',
@@ -81,14 +216,21 @@ const BARE_PLAN_DEFAULTS = {
 };
 
 function fixBareFrontmatter(file, src) {
-  if (!file.includes('docs/superpowers/plans/')) return src;
+  const rel = relative(ROOT, file);
+  if (
+    !rel.includes('docs/superpowers/plans/') &&
+    !rel.includes('docs/ephemeral/plans/')
+  )
+    return src;
   if (src.startsWith('---\n') || src.startsWith('---\r\n')) return src;
   for (const [stem, title] of Object.entries(BARE_PLAN_DEFAULTS)) {
     if (!file.includes(stem)) continue;
     const fm =
       `---\n` +
       `title: '${title}'\n` +
-      `status: delivered\n` +
+      `status: active\n` +
+      `purpose: build\n` +
+      `tier: ephemeral\n` +
       `date: 2026-04-17\n` +
       `---\n\n`;
     change(file, 'added frontmatter block');
@@ -97,14 +239,49 @@ function fixBareFrontmatter(file, src) {
   return src;
 }
 
+// Fix 6: backfill purpose + tier based on path heuristics
+function fixPurposeTier(file, src) {
+  const rel = relative(ROOT, file);
+  const m = src.match(/^(---\r?\n)([\s\S]*?)(\r?\n---\r?\n)/);
+  if (!m) return src;
+  const [, open, body, close] = m;
+
+  const hasPurpose = /^purpose:\s*\S/m.test(body);
+  const hasTier = /^tier:\s*\S/m.test(body);
+
+  if (hasPurpose && hasTier) return src; // already set
+
+  const inferred = inferPurposeTier(rel);
+  if (!inferred) return src; // no heuristic for this path
+
+  let newBody = body;
+  let changed = false;
+
+  if (!hasPurpose) {
+    newBody = `purpose: ${inferred.purpose}\n` + newBody;
+    change(file, `backfilled purpose: ${inferred.purpose}`);
+    changed = true;
+  }
+  if (!hasTier) {
+    newBody = `tier: ${inferred.tier}\n` + newBody;
+    change(file, `backfilled tier: ${inferred.tier}`);
+    changed = true;
+  }
+
+  return changed ? open + newBody + close + src.slice(m[0].length) : src;
+}
+
 const files = walk(DOCS);
 let writes = 0;
 for (const f of files) {
   const orig = readFileSync(f, 'utf8');
   let next = orig;
   next = fixStatusCasing(f, next);
+  next = fixStatusAlias(f, next);
+  next = fixAudienceAlias(f, next);
   next = fixAdr024Anomaly(f, next);
   next = fixBareFrontmatter(f, next);
+  next = fixPurposeTier(f, next);
   if (next !== orig) {
     writes++;
     if (APPLY) writeFileSync(f, next);
diff --git a/scripts/docs-frontmatter-schema.mjs b/scripts/docs-frontmatter-schema.mjs
index 1350dd6a9..4e8c0e876 100644
--- a/scripts/docs-frontmatter-schema.mjs
+++ b/scripts/docs-frontmatter-schema.mjs
@@ -1,93 +1,144 @@
 // Schema for docs/** frontmatter. Single source of truth.
-// Values reflect the current corpus minus drift (casing collapsed).
-// Extend here when a new legitimate value is introduced; avoid widening
-// to absorb mistakes — the point of the enum is to catch drift.
+// ADR-083 (2026-05-16): collapsed to 8 purposes × 4 tiers.
+//
+// STATUS collapsed from 22 → 4 canonical values.
+// CATEGORY replaced by free-form `topic` tags.
+// AUDIENCE collapsed from 14 → 3 canonical values.
+// PURPOSE and TIER are new required-for-new-docs enums.
+// STATUS_ALIAS_MAP and AUDIENCE_ALIAS_MAP support the transitional cycle.
 
-export const STATUS = [
-  'draft',
-  'active',
-  'accepted',
-  'in-progress',
+// === Canonical enums (new schema) ===
+
+export const PURPOSE = [
+  'orient',
+  'decide',
   'design',
-  'delivered',
-  'stable',
-  'deferred',
-  'superseded',
-  'archived',
-  'reference',
-  'template',
-  'raw',
-  'review',
-  'living',
+  'build',
+  'system',
+  'constrain',
+  'agent-context',
+  'remember',
 ];
 
-export const CATEGORY = [
-  'ai',
-  'analysis',
-  'architecture',
-  'compliance',
-  'components',
-  'data',
-  'design-spec',
-  'implementation',
-  'learning',
-  'living-index',
-  'methodology',
-  'navigation',
-  'patterns',
-  'reference',
-  'strategy',
-  'tutorial',
-  'workflow',
-];
+export const TIER = ['stable', 'living', 'ephemeral', 'card'];
 
-export const AUDIENCE = [
-  'analyst',
-  'developer',
-  'engineer',
-  'business',
-  'architect',
-  'admin',
-  'designer',
-  'product',
-  'manager',
-  'compliance',
-  'procurement',
-  'infosec',
-  'quality-manager',
-  'auditor',
-];
+export const STATUS = ['draft', 'active', 'superseded', 'archived'];
+
+// Free-form topic tags replace CATEGORY enum.
+// Dominant tags from the 2026-05-16 audit (for documentation; not enforced):
+//   adr, stats, charts, ux, projects, investigation, canvas, ax, coscout,
+//   methodology, azure, capability, response-paths, stores, marketplace, i18n, tier-gating
+// Any kebab-case value is valid.
+
+export const AUDIENCE = ['human', 'agent', 'both'];
+
+// === Alias maps for the transitional cycle (old → new) ===
+// Old values warn but pass; after Play 2 they hard-fail.
+
+export const STATUS_ALIAS_MAP = {
+  accepted: 'active',
+  'in-progress': 'active',
+  delivered: 'active',
+  reference: 'active',
+  template: 'active',
+  stable: 'active',
+  living: 'active',
+  review: 'draft',
+  design: 'draft',
+  raw: 'draft',
+  deferred: 'superseded',
+};
 
+export const AUDIENCE_ALIAS_MAP = {
+  developer: 'human',
+  engineer: 'human',
+  analyst: 'human',
+  business: 'human',
+  architect: 'human',
+  admin: 'human',
+  designer: 'human',
+  product: 'human',
+  manager: 'human',
+  compliance: 'human',
+  procurement: 'human',
+  infosec: 'human',
+  'quality-manager': 'human',
+  auditor: 'human',
+};
+
+// === Schema rules ===
 // Per doc-type rules. `general` covers docs/** except ADRs and specs.
-// ADRs have a permissive rule because historical conventions vary
-// (title-only is legal; adr-069 uses full general shape; adr-067 uses
-// title+status+date). Enforcing the full shape retroactively is a
-// separate ticket.
+// ADRs have a permissive rule because historical conventions vary.
+// Enforcing the full shape retroactively happens via docs-frontmatter-fix.mjs.
+
 export const schema = {
   general: {
-    required: ['title', 'audience', 'category', 'status'],
-    optional: ['related', 'last-reviewed', 'date'],
-    enums: { status: STATUS, category: CATEGORY, audience: AUDIENCE },
+    required: ['title'],
+    optional: [
+      'purpose',
+      'tier',
+      'status',
+      'audience',
+      'topic',
+      'related',
+      'last-verified',
+      'verified-against-commit',
+      'supersedes',
+      'date',
+    ],
+    enums: { status: STATUS, audience: AUDIENCE, purpose: PURPOSE, tier: TIER },
   },
   adr: {
     required: ['title'],
-    optional: ['status', 'date', 'related', 'audience', 'category', 'last-reviewed', 'supersedes', 'superseded-by'],
-    enums: { status: STATUS, category: CATEGORY, audience: AUDIENCE },
+    optional: [
+      'status',
+      'purpose',
+      'tier',
+      'date',
+      'related',
+      'audience',
+      'topic',
+      'last-verified',
+      'verified-against-commit',
+      'supersedes',
+      'superseded-by',
+    ],
+    enums: { status: STATUS, audience: AUDIENCE, purpose: PURPOSE, tier: TIER },
   },
   spec: {
     required: ['title', 'status'],
-    optional: ['audience', 'category', 'related', 'last-reviewed', 'date', 'type'],
-    enums: { status: STATUS, category: CATEGORY, audience: AUDIENCE },
+    optional: [
+      'purpose',
+      'tier',
+      'audience',
+      'topic',
+      'related',
+      'last-verified',
+      'verified-against-commit',
+      'supersedes',
+      'date',
+      'type',
+    ],
+    enums: { status: STATUS, audience: AUDIENCE, purpose: PURPOSE, tier: TIER },
   },
 };
 
+// === Path classifier ===
+// Updated for Play 1 new folder structure.
+// Old paths are kept for backward-compat during transition.
+
 export function classify(relPath) {
+  // New structure (Play 1)
+  if (relPath.startsWith('docs/living/decide/')) return 'adr';
+  if (relPath.startsWith('docs/living/design/specs/')) return 'spec';
+
+  // Legacy paths (still valid during transition)
   if (relPath.startsWith('docs/07-decisions/')) return 'adr';
   if (relPath.startsWith('docs/archive/adrs/')) return 'adr';
-  // Specs, plans, and other superpowers artifacts share the lighter
-  // required-fields rule (title + status) — they aren't published docs.
+  // Specs, plans, and other superpowers artifacts share the lighter rule.
   if (relPath.startsWith('docs/superpowers/')) return 'spec';
   if (relPath.startsWith('docs/archive/specs/')) return 'spec';
   if (relPath.startsWith('docs/archive/plans/')) return 'spec';
+
   return 'general';
 }
diff --git a/scripts/docs/gen-arch.mjs b/scripts/docs/gen-arch.mjs
new file mode 100644
index 000000000..0968baa3b
--- /dev/null
+++ b/scripts/docs/gen-arch.mjs
@@ -0,0 +1,288 @@
+#!/usr/bin/env node
+// gen-arch.mjs — auto-generate docs/05-technical/architecture-generated.md
+//
+// Reads:
+//   - packages/*/package.json + apps/*/package.json  → dependency graph (mermaid)
+//   - packages/*/tsconfig.json + apps/*/tsconfig.json → sub-path export/path map
+//   - apps/*/src/index.css                            → Tailwind @source directives
+//
+// Emits:
+//   docs/05-technical/architecture-generated.md
+//
+// Usage: node scripts/docs/gen-arch.mjs
+// Wire as: pnpm docs:gen-arch
+
+import { readFileSync, writeFileSync, readdirSync, statSync, existsSync } from 'node:fs';
+import { join, relative, dirname } from 'node:path';
+import { fileURLToPath } from 'node:url';
+
+const __dirname = dirname(fileURLToPath(import.meta.url));
+const ROOT = join(__dirname, '..', '..');
+const OUT = join(ROOT, 'docs', '05-technical', 'architecture-generated.md');
+
+const today = new Date().toISOString().slice(0, 10);
+
+// ---------------------------------------------------------------------------
+// Helpers
+// ---------------------------------------------------------------------------
+
+function readJson(path) {
+  try {
+    return JSON.parse(readFileSync(path, 'utf8'));
+  } catch {
+    return null;
+  }
+}
+
+function readText(path) {
+  try {
+    return readFileSync(path, 'utf8');
+  } catch {
+    return null;
+  }
+}
+
+/** List direct children of dir that are directories */
+function subdirs(dir) {
+  if (!existsSync(dir)) return [];
+  return readdirSync(dir)
+    .filter((entry) => {
+      try {
+        return statSync(join(dir, entry)).isDirectory();
+      } catch {
+        return false;
+      }
+    })
+    .map((entry) => join(dir, entry));
+}
+
+// ---------------------------------------------------------------------------
+// 1. Workspace discovery
+// ---------------------------------------------------------------------------
+
+const PACKAGE_DIRS = subdirs(join(ROOT, 'packages'));
+const APP_DIRS = subdirs(join(ROOT, 'apps'));
+
+// ---------------------------------------------------------------------------
+// 2. Dependency graph
+// ---------------------------------------------------------------------------
+
+function buildDepGraph() {
+  const nodes = [];
+  for (const dir of [...PACKAGE_DIRS, ...APP_DIRS]) {
+    const pkgJson = readJson(join(dir, 'package.json'));
+    if (!pkgJson) continue;
+    const name = pkgJson.name;
+    if (!name) continue;
+
+    const allDeps = {
+      ...pkgJson.dependencies,
+      ...pkgJson.devDependencies,
+      ...pkgJson.peerDependencies,
+    };
+
+    const workspaceDeps = Object.keys(allDeps).filter((d) => d.startsWith('@variscout/'));
+    const rel = relative(ROOT, dir);
+    nodes.push({ name, rel, deps: workspaceDeps });
+  }
+  return nodes;
+}
+
+/** Convert @variscout/foo-bar → FooBar (safe mermaid node ID) */
+function toNodeId(name) {
+  return name
+    .replace('@variscout/', '')
+    .split('-')
+    .map((part) => part.charAt(0).toUpperCase() + part.slice(1))
+    .join('');
+}
+
+function renderDepGraph(nodes) {
+  const lines = ['graph LR'];
+
+  const seen = new Set();
+  for (const node of nodes) {
+    const id = toNodeId(node.name);
+    if (!seen.has(id)) {
+      const label = node.name.replace('@variscout/', '');
+      const isApp = node.rel.startsWith('apps/');
+      const shape = isApp ? `["${label} (app)"]` : `["${label}"]`;
+      lines.push(`  ${id}${shape}`);
+      seen.add(id);
+    }
+  }
+
+  lines.push('');
+
+  const edgeSeen = new Set();
+  for (const node of nodes) {
+    const fromId = toNodeId(node.name);
+    for (const dep of node.deps) {
+      const toId = toNodeId(dep);
+      const edgeKey = `${fromId}->${toId}`;
+      if (!edgeSeen.has(edgeKey)) {
+        lines.push(`  ${fromId} --> ${toId}`);
+        edgeSeen.add(edgeKey);
+      }
+    }
+  }
+
+  return lines.join('\n');
+}
+
+// ---------------------------------------------------------------------------
+// 3. Sub-path export map
+// ---------------------------------------------------------------------------
+
+function buildExportMap() {
+  const rows = [];
+  for (const dir of [...PACKAGE_DIRS, ...APP_DIRS]) {
+    const pkgJson = readJson(join(dir, 'package.json'));
+    if (!pkgJson) continue;
+    const name = pkgJson.name;
+    if (!name) continue;
+
+    const exportsField = pkgJson.exports ?? {};
+    let exportKeys;
+    if (typeof exportsField === 'string') {
+      exportKeys = ['.'];
+    } else {
+      exportKeys = Object.keys(exportsField).sort();
+    }
+
+    const tsconfigPath = join(dir, 'tsconfig.json');
+    const tsconfig = readJson(tsconfigPath);
+    const pathKeys = Object.keys(tsconfig?.compilerOptions?.paths ?? {}).sort();
+
+    if (exportKeys.length === 0 && pathKeys.length === 0) continue;
+
+    rows.push({
+      pkg: name,
+      rel: relative(ROOT, dir),
+      exports: exportKeys,
+      paths: pathKeys,
+    });
+  }
+  return rows;
+}
+
+function renderExportMap(rows) {
+  const lines = [];
+  lines.push('| Package | `package.json` exports | `tsconfig.json` paths |');
+  lines.push('|---------|------------------------|----------------------|');
+
+  for (const row of rows) {
+    const exportsCell =
+      row.exports.length > 0
+        ? row.exports.map((e) => `\`${e}\``).join(', ')
+        : '_(none)_';
+    const pathsCell =
+      row.paths.length > 0
+        ? row.paths.map((p) => `\`${p}\``).join(', ')
+        : '_(none)_';
+    lines.push(`| \`${row.pkg}\` | ${exportsCell} | ${pathsCell} |`);
+  }
+
+  return lines.join('\n');
+}
+
+// ---------------------------------------------------------------------------
+// 4. Tailwind @source coverage
+// ---------------------------------------------------------------------------
+
+const SOURCE_DIRECTIVE_RE = /@source\s+"([^"]+)"/g;
+
+function buildTailwindMap() {
+  const rows = [];
+  for (const dir of APP_DIRS) {
+    const cssPath = join(dir, 'src', 'index.css');
+    const src = readText(cssPath);
+    const appName = readJson(join(dir, 'package.json'))?.name ?? relative(ROOT, dir);
+    const directives = [];
+    if (src) {
+      let match;
+      const re = /@source\s+"([^"]+)"/g;
+      while ((match = re.exec(src)) !== null) {
+        directives.push(match[1]);
+      }
+    }
+    rows.push({ app: appName, rel: relative(ROOT, dir), directives, hasCss: src !== null });
+  }
+  return rows;
+}
+
+function renderTailwindMap(rows) {
+  const lines = [];
+  lines.push('| App | `@source` directives in `src/index.css` |');
+  lines.push('|-----|------------------------------------------|');
+
+  for (const row of rows) {
+    if (!row.hasCss || row.directives.length === 0) {
+      const note = !row.hasCss ? '_(no src/index.css — not a Tailwind app)_' : '_(no @source directives)_';
+      lines.push(`| \`${row.app}\` | ${note} |`);
+    } else {
+      const cell = row.directives.map((d) => `\`${d}\``).join('<br>');
+      lines.push(`| \`${row.app}\` | ${cell} |`);
+    }
+  }
+
+  return lines.join('\n');
+}
+
+// ---------------------------------------------------------------------------
+// 5. Assemble document
+// ---------------------------------------------------------------------------
+
+const depNodes = buildDepGraph();
+const exportRows = buildExportMap();
+const tailwindRows = buildTailwindMap();
+
+const mermaidGraph = renderDepGraph(depNodes);
+const exportTable = renderExportMap(exportRows);
+const tailwindTable = renderTailwindMap(tailwindRows);
+
+const doc = `---
+title: 'Architecture (auto-generated)'
+purpose: system
+tier: living
+audience: agent
+topic: [architecture, generated]
+status: active
+last-verified: ${today}
+---
+
+> AUTO-GENERATED by \`pnpm docs:gen-arch\` — do not edit by hand. Regenerate after changing \`package.json#dependencies\`, \`tsconfig.json#paths\`, \`package.json#exports\`, or \`apps/*/src/index.css\` \`@source\` directives.
+
+# Architecture (auto-generated)
+
+Generated: ${today}. Source: \`scripts/docs/gen-arch.mjs\`.
+
+---
+
+## Workspace Dependency Graph
+
+Internal workspace dependencies only (edges within the \`@variscout/*\` namespace). External npm dependencies omitted for clarity.
+
+\`\`\`mermaid
+${mermaidGraph}
+\`\`\`
+
+---
+
+## Sub-Path Export Map
+
+Sub-path exports declared in \`package.json#exports\` and corresponding TypeScript path aliases in \`tsconfig.json#compilerOptions.paths\`. These must be updated together — adding one without the other silently breaks imports. See \`.claude/INVARIANTS.md\` §Sub-path exports.
+
+${exportTable}
+
+---
+
+## Tailwind v4 @source Coverage
+
+Each app must declare \`@source\` directives for every shared package whose class names it uses. Missing a directive silently breaks Tailwind v4 responsive utilities. See \`.claude/INVARIANTS.md\` §\`@source\` directive.
+
+${tailwindTable}
+`;
+
+writeFileSync(OUT, doc, 'utf8');
+console.log(`docs:gen-arch: wrote ${OUT}`);