chore(deps): bump actions/upload-artifact from 4.6.2 to 7.0.0#2
Closed
dependabot[bot] wants to merge 1 commit into
Closed
chore(deps): bump actions/upload-artifact from 4.6.2 to 7.0.0#2dependabot[bot] wants to merge 1 commit into
dependabot[bot] wants to merge 1 commit into
Conversation
dependabot
Bot
added
dependencies
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact) from 4.6.2 to 7.0.0. - [Release notes](https://github.com/actions/upload-artifact/releases) - [Commits](actions/upload-artifact@ea165f8...bbbca2d) --- updated-dependencies: - dependency-name: actions/upload-artifact dependency-version: 7.0.0 dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>
dependabot
Bot
force-pushed
the
dependabot/github_actions/actions/upload-artifact-7.0.0
branch
from
4057353 to
a58ec07
Compare
Owner
|
Applied this update directly on main in a consolidated dependency bump commit (all 13 Dependabot updates at once). Closing this PR. |
Contributor
Author
|
OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting If you change your mind, just re-open this PR and I'll resolve any conflicts on it. |
dependabot
Bot
deleted the
dependabot/github_actions/actions/upload-artifact-7.0.0
branch
jukka-matti
added a commit
that referenced
this pull request
Apr 26, 2026
Defers #2 (full i18n sweep) — that spans the whole apps/azure app, not just process hub; track separately. - processHub.ts: orphan/synthesized hubs now get a friendly fallback name ('Unknown hub') instead of rendering the raw hubId. Applies to both investigation-derived and evidence-snapshot-derived synthesized hubs. - ProcessHubEvidencePanel.tsx: false-green severity is now rate-based — 0 false greens → green, ≤1% rate → amber, >1% → red. Replaces the hardcoded red-for-any-count default. Threshold (FALSE_GREEN_AMBER_RATE) exposed as a constant; Phase 6 spec can refine. - @variscout/core/i18n: new formatPlural() helper using Intl.PluralRules. Applied to ProcessHubCard + ProcessHubReviewPanel (replacing hand-rolled "investigation"+"s", formatChangeSignals, formatOverdueActions). Helper is locale-aware (Polish 'few' tested) and required-other-fallback safe. - ProcessHubReviewPanel.tsx: split from 549 LOC into 4 focused files: ProcessHubReviewPanel.tsx 106 LOC (orchestrator) ProcessHubCadenceQuestions.tsx 37 LOC (3 cadence question bands) ProcessHubCadenceQueues.tsx 363 LOC (daily + weekly columns) ProcessHubFormat.ts 93 LOC (shared format helpers) Tests stay green (16/16) — pure structural refactor. Co-Authored-By: ruflo <ruv@ruv.net>
This was referenced Apr 27, 2026
jukka-matti
added a commit
that referenced
this pull request
Apr 27, 2026
Lift the Azure-only ProcessHubCurrentStatePanel into @variscout/ui so Phase 2 snapshot-mode work in LayeredProcessView can compose with it from a shared home. Component is fully props-based — takes CurrentProcessState in, renders JSX out — no Azure-specific deps (Blob Storage, EasyAuth, app stores). The two formatter dependencies (formatMetric, formatChangeSignals) that previously lived in Azure's ProcessHubFormat helper are inlined here using the canonical formatStatistic + formatPlural from @variscout/core/i18n. Adds 8 unit tests covering: heading + severity badge, lens count cards, empty state, +N indicator at >6 items, Cpk-vs-target detail, change-signal pluralization, detail fallback, response-path label. PWA does not get this panel — Process Hub is an Azure-only feature (multi-investigation aggregation requiring persistence per ADR-072); PWA is session-only by ADR-012. Confirmed during planning. Phase 2 PR #2 of N. Co-authored-by: ruflo <ruv@ruv.net>
jukka-matti
added a commit
that referenced
this pull request
Apr 27, 2026
Combines three sequenced PRs into one spec per feedback_full_vision_spec memory: - PR #1 (H1): team notes on state items — question / gemba / data-gap / decision kinds, persisted in ProcessHubInvestigationMetadata.stateNotes - PR #2 (V2 follow-up): full EvidenceSheet rendering — refactors ProcessHubEvidenceContract to split countFor (sync) + loadFindingsFor (async), eliminates the synthetic-Finding cast from PR #99 - PR #3 (H2 launch): General Evidence Source for CSV/Excel — adds GENERIC_TABULAR_PROFILE to DATA_PROFILE_REGISTRY, mapping confirmation UI in ProcessHubEvidencePanel, cadence metadata Resolves both prior open questions inline: - Evidence contract split (countFor + loadFindingsFor + onChipClick + onFindingSelect — cleaner than overloading findingsFor sync/async) - Mapping UX explicit confirmation (matches operating-model § "analyst confirms mappings" language) Updates spec index, marks the parent V2 actionable-panel spec as Delivered (PRs #98 #99). Co-Authored-By: ruflo <ruv@ruv.net>
jukka-matti
added a commit
that referenced
this pull request
Apr 27, 2026
Sequenced TDD breakdown for 3 PRs: - PR #1: team notes on state items (~250-350 LOC, 8 tasks). New ProcessStateNote types in core, ProcessHubInvestigationMetadata extension, ProcessHubNotesContract on the panel, StateItemNotesDrawer in apps/azure, Dashboard wiring via useStorage round-trip. - PR #2: full EvidenceSheet rendering (~300-400 LOC, 5 tasks). Refactors ProcessHubEvidenceContract to split countFor (sync) + loadFindingsFor (async); EvidenceSheet bottom sheet; Dashboard owns sheet state + async load via useStorage; eliminates synthetic-Finding cast from PR #99. - PR #3: General Evidence Source for CSV/Excel (~600-800 LOC, 4 tasks). GENERIC_TABULAR_PROFILE in DATA_PROFILE_REGISTRY; profile picker UI; mapping confirmation form; cadence selector (5 actual EvidenceCadence values); telemetry on save. Three implementation reality notes flagged at top: 1. Actual DataProfileDefinition signature differs from spec sketch (single 'label' field; detect takes only rows; confidence is 'high'/'medium'/'low' enum not number). 2. EvidenceCadence enum lacks 'monthly' — use existing 5 values. 3. Spec back-link upgrades to Markdown link in PR #1 Task 7. Co-Authored-By: ruflo <ruv@ruv.net>
5 tasks
jukka-matti
added a commit
that referenced
this pull request
Apr 28, 2026
…d telemetry
The Phase 3 spec's telemetry table specified payload {hubId, responsePath,
lens, evidenceCount} but the implementation omitted evidenceCount.
evidenceCount is the key engagement metric for the chip feature.
Threads count through the panel's onChipClick(item, hubId, count) so
Dashboard can include it in safeTrackEvent without re-deriving.
Phase 3 PR #2, code-review followup.
Co-Authored-By: ruflo <ruv@ruv.net>
jukka-matti
added a commit
that referenced
this pull request
Apr 28, 2026
* refactor(ui): split ProcessHubEvidenceContract into countFor + loadFindingsFor Replaces the synthetic-Finding placeholder pattern from PR #99 with a clean separation of concerns: - countFor (sync) — chip badge count, derived from rollup metadata - loadFindingsFor (async) — fetches real findings for the sheet, only called when consumer wires it after chip click - onChipClick(item) — telemetry signal, no findings arg - onFindingSelect(item, finding) — sheet's selection callback Sheet management stays on the consumer side (next task wires Dashboard). Eliminates the 'as unknown as readonly Finding[]' cast and the synthetic- placeholder generation entirely. Existing chip tests rewritten for the new contract; 2 new tests added (no-load-on-render guard + onFindingSelect contract presence). Phase 3 PR #2, Task 2. Co-Authored-By: ruflo <ruv@ruv.net> * feat(azure): add EvidenceSheet bottom sheet for finding click-thru Lightweight sheet listing findings (label + status badge + click-thru). Renders nothing when item is null; loading state when findings is null; empty state for zero findings. Esc key, click-outside, and close button all fire onClose. Phase 3 PR #2, Task 3. Co-Authored-By: ruflo <ruv@ruv.net> * feat(azure): wire EvidenceSheet with lazy finding load on chip click Replaces PR #99's synthetic-Finding placeholder pattern with a real async load through the existing useStorage().loadProject() chain. ProcessHubReviewPanel: - countFor derives chip count from ProcessHubInvestigationMetadata.findingCounts (same arithmetic as the previous synthetic-placeholder length) - loadFindingsForItem / onChipClick / onFindingSelect pass-through to Dashboard - removes the synthetic Finding[] cast and TODO(evidence-sheet-pr) comment - removes safeTrackEvent import (no longer needed in this component) Dashboard: - sheetState managed at Dashboard level (item + hubId + findings | null) - handleChipClick: fires telemetry, opens sheet in loading state, async-loads via loadProject() chain, applies findings with a race guard - handleFindingSelect: fires telemetry, closes sheet, navigates via existing onOpenProject + best-effort URL hash for finding deep-link - EvidenceSheet rendered at Dashboard level (sibling of StateItemNotesDrawer) Telemetry: process_hub.evidence_sheet_{opened,finding_clicked} with non-PII payloads (hubId from rollup.hub.id, lens, responsePath, findingStatus enum in finding-click event). No PII per ADR-059. Phase 3 PR #2, Task 4. Co-Authored-By: ruflo <ruv@ruv.net> * fix(azure): drop dead URL-hash guard in handleFindingSelect The guard `window.location.hash !== '#finding-${id}'` (with `#`) compared against an assignment without `#` — the comparison was always false so the guard never short-circuited. Assignment is idempotent in browsers anyway; just drop the guard. Phase 3 PR #2, code-review followup. Co-Authored-By: ruflo <ruv@ruv.net> * fix(azure): include evidenceCount in process_hub.evidence_sheet_opened telemetry The Phase 3 spec's telemetry table specified payload {hubId, responsePath, lens, evidenceCount} but the implementation omitted evidenceCount. evidenceCount is the key engagement metric for the chip feature. Threads count through the panel's onChipClick(item, hubId, count) so Dashboard can include it in safeTrackEvent without re-deriving. Phase 3 PR #2, code-review followup. Co-Authored-By: ruflo <ruv@ruv.net> --------- Co-authored-by: ruflo <ruv@ruv.net>
jukka-matti
added a commit
that referenced
this pull request
May 4, 2026
…g) (#121) * docs: mark layered-process-view-v1 plan delivered V1 component, FrameView swaps (PWA + Azure), and doc updates (ADR-070, methodology.md, mental-model-hierarchy.md, llms.txt) all shipped via production-line-glance C2 (merged 2026-04-29). Plan frontmatter was never promoted; flipping deferred → delivered. Co-Authored-By: ruflo <ruv@ruv.net> * docs: consolidate product vision into one canonical spec Adds docs/superpowers/specs/2026-05-03-variscout-vision-design.md as the canonical product vision. Supersedes the 2026-04-27 operating-model and product-method-roadmap specs (moved to docs/archive/specs/ with status: superseded + forward pointer). methodology.md gets a forward-pointer banner and survives as longer-form companion. All inbound + outbound cross-refs updated; specs/index.md rows show status: Superseded. Vision thesis: "the map is the product" — Process Hub IS its logic map; one continuous canvas (DAG with branch + join + two-level nesting + context propagation) replaces the FRAME river-SIPOC + LayeredProcessView + LayeredProcessViewWithCapability stack; cards-with-mini-charts per step + drill-down panel + mode lenses replace the separate Analysis tab; "tributary" / "CTS" jargon retired. Engine and data model survive. 10 canvas commitments in spec section 3.3 are load-bearing. 11 open questions in section 8 carry brainstorm defaults that need explicit confirmation before implementation plans are written. Decision-log pinned under Replayed Decisions (2026-05-03). In-flight plan reconciliation (frontmatter status flips): - delivered: production-line-glance C1/C2/charts (PRs #105/#106/#107), actionable-current-process-state-panel (PR #98), phase-3 H1/H2 (PRs #100/#101/#102) - in-progress (corrected): coscout-intelligence-architecture, question-driven-eda-2-handoff, multi-level-scout-v1 - unchanged: phase-6-implementation-plan (deferred), investigation-wall (in-progress) Brainstorm transcript: ~/.claude/plans/i-would-like-to-composed-rose.md. Co-Authored-By: ruflo <ruv@ruv.net> * docs: lock vision §8 decisions; canvas replaces Frame+Analysis tabs Walk-through resolved 11 brainstorm-default markers in the 2026-05-03 vision spec §8, plus a new Q0 ("tab vs canvas scaffold") that emerged during journey mapping. Headlines: - Q0: Canvas eats Frame + Analysis tabs. Investigation / Improvement / Report keep their own surfaces. Wall is dual-home (destination + canvas overlay). JourneyPhase retires as a top-level nav primitive. - Q1: Drill-down is a modern floating overlay anchored to the clicked card with blurred-canvas backdrop (overrides spec's "side-panel slides from right" default — that conflicted with the locked C3 supersession of CoScout's right-rail claim). - Q5: No CoScout map drafting in V1; manual click/drag/connect canvas authoring is the path (overrides spec default). - Q7: Hard cutover, no migration window — no users yet to preserve. - Q8: PWA gets local Hub-of-one with IndexedDB persistence; Azure adds cloud sync + multi-Hub + cadence + CoScout + team features. - Q4 + Q11: Wall is dual-home; promoted hypotheses render as node markers, drafts as faint arrows. Vision spec: §3.4, §5.2, §5.3, §5.4, §5.6, §5.7, §6, §7, §8 rewritten in place; §8 replaced "open questions" with a "resolved decisions" table; frontmatter status promoted draft → accepted. ADR-070 (FRAME workspace): added 2026-05-03 supersession amendment explicitly retiring FRAME as a top-level route; lists what survives inside the Canvas vs what gets deleted in the same PR that ships it. Decision-log: appended 2026-05-03 entry summarizing all 12 decisions + pointer to the §8 walkthrough plan. Full rationale + spec/code follow-ups in ~/.claude/plans/lets-do-this-next-rustling-simon.md. Co-Authored-By: ruflo <ruv@ruv.net> * docs: §8 follow-ups — glossary, ADR amendments, roadmap re-tag Smaller doc follow-ups triggered by the 2026-05-03 vision §8 resolution (see commit f12e8b1 for the headline changes). - docs/glossary.md (Q10): canonicalized as the single home for VariScout terminology. New "Process methodology terms" section (step / sub-step / column / input / output / outcome / branch & join / Process Hub) and a "Retired terms" table (tributary → factor; CTS → outcome; FRAME workspace → Canvas; Analysis tab → Canvas + lenses; Layered Process View → Canvas with lenses; etc) with replacements + retirement reasons. - docs/01-vision/methodology.md (Q10): added forward pointer to glossary.md as the canonical terminology home; methodology stays as the longer narrative companion but stops re-defining terms. - ADR-068 (Q3): added 2026-05-03 amendment formalizing modes and levels as orthogonal axes. Modes = which lens; levels = which slice (canvas pan/zoom). Level inferred from canvas state, not a separate picker. Refines (does not replace) the 2026-04-28 level-aware coaching amendment. - ADR-059 (Q8): added 2026-05-03 amendment specifying PWA local Hub-of-one with IndexedDB persistence. Browser-tenant-only — Constitution P1 (browser-only processing) and P8 (no AI in free tier) preserved. Schema scope listed for the FRAME canvas detail spec to inherit. - product-method-roadmap-design.md (Q6): added explicit "this is sequencing, not destination" header. Status was already superseded; the new framing tells stakeholders to read the vision spec first and consult this doc only for delivery-order intent. docs:check green. No broken cross-refs, no orphans, frontmatter valid. Co-Authored-By: ruflo <ruv@ruv.net> * docs: add framing-layer design spec (Spec 1 of canvas-detail decomposition) First of five canvas-detail specs decomposed from the 2026-05-03 vision §8 walkthrough. Covers everything BEFORE canvas authoring proper — Hub creation, data ingestion, investigation entry, multi-source join, defect anchoring, Pareto integration, and the three composable canvas filter states. Two-layer model: - Hub-level (durable): process goal narrative + outcome measure(s) + specs + primary scope dimensions - Investigation-level (episodic): Issue → Question → CTP→CTQ Hypothesis → Evidence → Finding Mode B (first-time Hub creation): - Stage 1: free-text goal + scaffold chips (Purpose / Customer / What matters) - Stage 2: paste/upload (existing UX) - Stage 3: outcome confirmation with inline distribution + data quality + per-candidate inline specs (no σ-based suggestions — specs are customer-driven) + primary scope dimensions sub-step - Stage 5: investigation entry as floating modal with skip-path → observation-triggered EDA Mode A (reopening existing Hub): - A.1: reopen, no new data → straight to canvas - A.2-paste: match-summary card with timeline preview + 2-axis classifier (source × temporal); blocks on overlap and different-grain - A.2-evidence-source: background ingestion with "X new snapshots ↑" indicator (Azure only — PWA Hub-of-one has no cloud sync per Q8) Multi-source via shared keys: Hub spine = process line; multiple sources joined via lot_id / batch_id / part_id; per-source provenance; per-source independent timelines (V1). Defect data + Pareto: multi-anchor (per-step rejects + outcome); canvas-attached Pareto with two pickers (group-by × weighted-by); Y-axis adapts to active mode lens (count / cost / time / Cpk gap / %OOS / ...); Pareto-bar selection becomes canvas-wide scope filter with "Make this the investigation scope" promote-to-investigation. Three composable canvas filter states (time-window / scope / Pareto-grouped) compose declaratively without new aggregation primitives. ADR-073 holds: per-(node × context-tuple × scope) buckets stay heterogeneity-aware. Engine reuse: detectColumns() + inferMode() + validateData() + suggestNodeMappings + TimelineWindow + ParetoChart + StepErrorPareto + statusForCpk + investigation entities all already shipped. Almost no new infrastructure required. Spec is 638 lines, 16 sections including verification criteria. Indexed at docs/superpowers/specs/index.md. Inherits Q0 + Q1–Q11 anchors from §8 walkthrough at ~/.claude/plans/lets-do-this-next-rustling-simon.md. Sibling specs (deferred): Spec 2 manual canvas authoring, Spec 3 cards / drill-down / mode lenses, Spec 4 canvas overlays + Wall sync, Spec 5 IndexedDB persistence schema. Co-Authored-By: ruflo <ruv@ruv.net> * docs: revise Q8 — PWA persistence opt-in; .vrs files = shareable training scenarios Walking back the original Q8 wording ("PWA = local Hub-of-one with IndexedDB persistence" — auto-on) to Option 4 hybrid: session-only by default; opt-in via "Save to this browser" → IndexedDB-backed Hub-of-one; .vrs file export/import always available. Strategic reframe: PWA serves four personas (LSSGB students, demos, casual analysts, AND trainers authoring scenarios). Trainers package datasets + Hub state + sample investigations into a .vrs bundle and share via LMS / email; students import to start from a prepared training state. .vrs becomes the scenario-distribution format, positioning PWA as the methodology-teaching surface. Persistence is now explicit consent per persona — training students opt in once and auto-save; demo users skip; privacy-conscious users export to file; trainers export+share. Constitution P1 (browser-only processing) and P8 (no AI in free tier) preserved. Companies still use Azure tier for centralized + secure persistence per ADR-059. Updates: - vision spec §7 tier paragraph rewritten + §8 Q8 row updated - framing-layer spec §15 V1 scope adds opt-in Save-to-browser + .vrs export/import + IndexedDB-loaded-post-opt-in; §14 anchors table Q8 row revised - decision-log §1 entry added: "Q8 revised — PWA persistence opt-in" with strategic rationale (training scenarios; trainer persona; no surprise persistence on shared computers) - apps/pwa/CLAUDE.md hard rule updated from "no persistence" to "session-only by default; opt-in IndexedDB allowed; .vrs import/export for trainer-shared scenarios" docs:check green. Co-Authored-By: ruflo <ruv@ruv.net> * plan: Framing Layer V1 Slice 1 — Foundation (16 tasks, TDD bite-sized) First implementation plan from the framing-layer spec. Slice 1 scope (smallest end-to-end): - Mode B Stages 1–3 + canvas first paint (PWA + Azure) - Mode A.1 reopen (Azure always; PWA only if opted in) - PWA opt-in IndexedDB persistence (Q8-revised Option 4) - PWA .vrs file export/import (round-trip; trainer scenarios) - ProcessHub schema extension (processGoal, outcomes, primaryScopeDimensions) - Goal-context biased outcome detection - Multi-outcome validateData refactor - Inline per-candidate spec editor (no σ-based suggestions) - Characteristic-type-aware spec defaults - Primary scope dimensions auto-suggest + multi-select - Graceful degradation banner (no goal-keyword match) 16 tasks, TDD bite-sized (failing test → run + fail → implement → run + pass → commit). Task 0 locks 5 load-bearing decisions in a companion file. Subagent-driven default execution per feedback_subagent_driven_default; Sonnet workhorse. Out of scope (later slices): - Slice 2: Stage 5 modal + Mode A.2-paste + A.2-evidence-source - Slice 3: Multi-source via shared keys - Slice 4: Defect anchoring + Pareto on canvas Plan referenced from: - docs/decision-log.md §4 Session Backlog (queued row added) - docs/superpowers/specs/2026-05-03-framing-layer-design.md (Implementation plans section added — lists all 4 V1 slices) docs:check green; 471 docs frontmatter-valid. Co-Authored-By: ruflo <ruv@ruv.net> * plan: expand Tasks 6–15 to full TDD step-by-step Per writing-plans skill 'no placeholders' rule: each task in a plan must be self-contained because subagent-driven execution reads tasks out of order in fresh subagents per task. Original slice-1 plan had Tasks 1–5 fully step-by-step but Tasks 6–15 sketched ('follow the same TDD shape') — violated the rule. Expanded all 10 sketched tasks to the same TDD pattern as Tasks 1–5: failing test code → run + fail expected output → implementation code → run + pass expected output → commit message. Plan grew 1135 → 3272 lines. Tasks now self-contained: - 6: OutcomeCandidateRow (6 RTL tests) - 7: characteristicTypeDefaults (7 tests) - 8: suggestPrimaryDimensions + selector UI (9 tests) - 9: OutcomeNoMatchBanner (3 RTL tests) - 10: GoalBanner + OutcomePin (8 RTL tests) - 11: .vrs file format + export/import (5 tests) - 12: PWA Dexie hubRepository (7 tests) - 13: SaveToBrowserButton + VrsExport/Import (6 tests) - 14: PWA Mode B E2E + Mode A.1 RTL (3 tests) - 15: Azure HubCreationFlow + ProcessHubView GoalBanner (2 tests) Plan ready for superpowers:subagent-driven-development. docs:check green. Co-Authored-By: ruflo <ruv@ruv.net> * plan: lock 5 decisions for framing-layer V1 slice 1 D1: AnalysisBrief unchanged; new OutcomeSpec type D2: PWA persistence opt-in via 'Save to this browser' D3: validateData multi-outcome refactor D4: Goal-context biasing is deterministic keyword extraction D5: Mode A.1 reopen gated by IndexedDB opt-in flag (no PWA Hub list) * feat(core): add framing-layer fields to ProcessHub processGoal: string narrative (Hub-level durable) outcomes: OutcomeSpec[] (CTQs in plain language) primaryScopeDimensions: string[] (prominent picker dimensions) OutcomeSpec includes characteristicType (nominalIsBest / smallerIsBetter / largerIsBetter) — drives spec-input UI per framing-layer spec §5.3. Specs are customer-driven; no σ-based suggestions per Q8 / spec §3.3. * feat(core): extractHubName utility Extracts a short Hub name (≤50 chars, word-boundary truncated) from the first sentence of a goal narrative. Used by Stage 1 HubGoalForm to auto-populate the Hub name field on goal-narrative confirm. * fixup! feat(core): extractHubName utility Remove unnecessary .trimEnd() — plan's slice + lastIndexOf + final trim already handles the space-at-boundary case correctly without dropping an extra word. Co-Authored-By: ruflo <ruv@ruv.net> * feat(core): goal-context biasing in detectColumns Optional DetectColumnsOptions.goalContext threads the user's goal narrative through to outcome ranking. Tokenizes the narrative (lowercased, stopwords removed), computes per-column keyword overlap with the goal, adds a 0.5-weight bonus on top of existing scoring. Implements framing-layer spec §3.1: goal narrative biases deterministic detection at Stage 3. Deterministic — no AI (per Q5). Backward-compatible (option is optional). * feat(parser): refactor validateData to accept string[] outcomeColumns Signature changed from `(data, outcomeColumn: string | null)` to `(data, outcomeColumns: string[])`. A row is excluded when ANY listed outcome column is invalid. New `perOutcome: Record<string, PerOutcomeQuality>` field on DataQualityReport exposes per-column valid/invalid/missing counts. All six existing call sites updated; existing tests migrated to array API; no backward-compat overload (D3, feedback_no_backcompat_clean_architecture). Co-Authored-By: ruflo <ruv@ruv.net> * feat(ui): HubGoalForm component (framing layer Stage 1) Free-text textarea with three scaffold chips ('+ Purpose', '+ Customer', '+ What matters') that insert labeled prompt chunks into the narrative on click. Optional 'See examples' link surfaces three anonymized goal narratives. Skip-framing path supported. Implements framing-layer spec §5.1 — Option C 'free-text + scaffold chips' shape locked during brainstorm. * feat(ui): OutcomeCandidateRow with inline per-candidate specs Stage 3 horizontal candidate row layout: radio + name + sparkline + (when selected) inline spec inputs + quality stack + match confidence. LSL/USL placeholders say 'from customer spec' (no σ-based suggestions per Q8/Q2). LSL disabled for smallerIsBetter; USL disabled for largerIsBetter. Adds @variscout/core/processHub sub-path export so consumers can reach the new CharacteristicType / OutcomeSpec types without colliding with the unrelated CharacteristicType in core/types.ts. Implements framing-layer spec §5.3.1. Co-Authored-By: ruflo <ruv@ruv.net> * feat(core): characteristic-type-aware spec defaults inferOutcomeCharacteristicType: case-insensitive name-keyword detection (defect/reject/scrap/fail/error/fault → smallerIsBetter; yield/uptime/throughput/recovery/efficiency → largerIsBetter; default nominalIsBest). Renamed from the plan's inferCharacteristicType to avoid collision with the existing spec-limit-based inferCharacteristicType in core/types.ts. defaultSpecsFor: returns target=mean for nominalIsBest; target=0 for smallerIsBetter; no target for largerIsBetter. cpkTarget=1.33 (literature default) across the board. No LSL/USL suggestions — those are customer-driven (per Q8/Q2). 7 tests in packages/core/src/specs/__tests__/characteristicTypeDefaults.test.ts. Re-exported at root barrel as inferOutcomeCharacteristicType, defaultSpecsFor, DataStats. Co-Authored-By: ruflo <ruv@ruv.net> * feat: PrimaryScopeDimensions auto-suggest + selector UI suggestPrimaryDimensions filters columns by cardinality (3-50) + name keyword (id/lot/batch/product/customer/shift/operator/ machine/supplier/sku/plant/line/station). High-cardinality columns (>50 levels) flagged as 'join key, not Pareto candidate'. PrimaryScopeDimensionsSelector pre-checks suggested columns; user confirms or overrides; skippable. Implements framing- layer spec §3.4 and §5.3.2. Co-Authored-By: ruflo <ruv@ruv.net> * feat(ui): OutcomeNoMatchBanner graceful degradation Renders when max(candidate.matchScore) < 0.30 (cryptic columns or vague goal). Surfaces rename affordance, free-text 'I expected the outcome to be' note, and skip-outcome path. Hub records the user's pick for future detection (slice 1 stub; full learning loop in Spec 5). Implements framing-layer spec §5.3.3. * feat(ui): GoalBanner + OutcomePin canvas first-paint components GoalBanner renders Hub.processGoal at top of canvas; click-to-edit opens textarea with Save / Cancel; onChange callback to parent. Hides itself when goal is empty. OutcomePin shows Cpk badge when specs are set + n>=10; falls back to 'mean ± σ + n' + '+ Add specs' chip when specs absent; shows 'Trust pending' when specs set but n<10. Per Q2 fallback rule (framing-layer spec §5.2) and sample-size honesty (vision §2.3). * feat(core): .vrs file format + export/import (round-trip tested) VRS_VERSION='1.0'. VrsFile shape: version, exportedAt, hub, optional rawData, optional metadata (exportSource pwa|azure + appVersion). vrsExport(hub, rawData?, metadata?) → JSON string (pretty, indent 2). vrsImport(json) → VrsFile with version validation + clear error messages on mismatch. Hub-only export supported (rawData optional, useful for canvas-without-data sharing). Round-trip preserves Hub + rawData via JSON serialization. .vrs files double as shareable training scenarios per Q8-revised. Co-Authored-By: ruflo <ruv@ruv.net> * feat(pwa): Dexie hubRepository for opt-in Hub-of-one persistence New PwaDatabase (Dexie v1) with two tables: hubs (single-row Hub-of-one constraint) + meta (opt-in flag). hubRepository API: getOptInFlag / setOptInFlag (auto-clears Hub on false) / saveHub (overwrites single row) / loadHub (returns null if absent) / clearHub. Per Q8-revised: Dexie loaded only after explicit opt-in via 'Save to this browser' affordance (Task 13). Default behavior unchanged from today (session-only). dexie + fake-indexeddb added as dependencies. Co-Authored-By: ruflo <ruv@ruv.net> * feat(pwa): Save-to-browser opt-in + .vrs export/import buttons SaveToBrowserButton: shows 'Save to this browser' when not opted in; clicking opts in + persists Hub. Once opted in, auto-saves on Hub change. Button switches to 'Saved · Forget' — click + confirm clears persistence (calls setOptInFlag(false) which auto-clears Hub). VrsExportButton: serializes current Hub + data via vrsExport, downloads as <hub-name>.vrs (sanitized). VrsImportButton: file picker; reads file → vrsImport → emits onImport({ hub, rawData }) for parent to consume. Per Q8-revised hybrid persistence (Option 4). * feat(pwa): Mode A.1 reopen wiring + SessionProvider scaffold SessionProvider holds current Hub + rawData in React Context (in-memory; persistence via opt-in hubRepository). App.tsx: on mount, getOptInFlag → if true, loadHub() and seed sessionHub. GoalBanner renders above main content when sessionHub.processGoal is set, so a Mode A.1 reopen surfaces the saved Hub goal immediately. RTL test verifies both branches: opt-in false → HomeScreen; opt-in true + saved Hub → canvas with goal banner restored from IndexedDB. Mode B end-to-end (paste → goal → mapping → canvas first paint with OutcomePin per outcome) requires the ColumnMapping refactor that is out of scope for slice 1; the Playwright spec is checked in as .skip to be re-enabled when Stage 1/3 routing lands inside App.tsx. * feat(pwa): inject HubGoalForm Stage 1 into Mode B paste flow After paste analyze, before ColumnMapping, render HubGoalForm so the user states the process goal. On confirm, store narrative in sessionStore. On skip, sentinel '' marks the Stage 1 step done without a narrative. ColumnMapping now waits for goalNarrative !== null. ColumnMapping internals unchanged — Stage 3 refactor (using OutcomeCandidateRow / PrimaryScopeDimensionsSelector / OutcomeNoMatchBanner) ships in slice 2. Hub.processGoal is set at ColumnMapping confirm if narrative was provided; outcomes[] and primaryScopeDimensions are left for slice 2 (TODO marker in handleMappingConfirmWithGoal). Co-Authored-By: ruflo <ruv@ruv.net> * feat(azure): mount GoalBanner above Hub tabs + Dexie v7 no-op ProcessHubView renders <GoalBanner goal={hub.processGoal}> above the existing tab container — Mode A.1 reopen surfaces the saved process goal immediately on Hub load (Azure tier persists Hub-level state via Dexie always; no opt-in needed). Dexie schema bumped 6 -> 7 as a no-op version; the new ProcessHub fields (processGoal, outcomes, primaryScopeDimensions) added in Task 1 are optional TypeScript additions that Dexie stores without schema migration. (Plan said v4 -> v5; the actual Azure schema was already at v6, so the bump lands at v7.) GoalBanner is mounted read-only for slice 1 because ProcessHubView does not currently receive a Hub-update callback in its props; wiring onChange is deferred to slice 2 alongside HubCreationFlow. HubCreationFlow component (Stage 1 -> Stage 2 -> Stage 3 -> built Hub) and the Dashboard "+ New Hub" button are deferred to slice 2 — the plan's Stage3Mapping implementation is sketch-only and the full ColumnMapping refactor required to wire OutcomeCandidateRow / PrimaryScopeDimensionsSelector / OutcomeNoMatchBanner is out of scope for slice 1 (mirrors the slice-1 PWA Task 14 follow-up). Co-Authored-By: ruflo <ruv@ruv.net> * docs: framing-layer spec status draft -> active Spec is the active working design. Slice 1 implements its foundation; Stage 3 ColumnMapping refactor + multi-source + defect anchoring continue across slices 2-4. Decision-log session-backlog row already reflects in-flight status (commit d49cbfe). * docs: update slice-1 row with delivery state + slice-2 deferrals Slice 1 row → in-flight with the foundation enumeration and explicit slice-2 deferrals (Stage 3 ColumnMapping refactor, canvas first-paint OutcomePin, button mounting in PWA UI, Azure HubCreationFlow, Mode B E2E). Per final-reviewer finding (Important #2): Save-to-browser + .vrs Export/Import buttons are tested + scaffolded but currently unmounted — slice 2 will surface them in the PWA chrome. --------- Co-authored-by: ruflo <ruv@ruv.net>
6 tasks
jukka-matti
added a commit
that referenced
this pull request
May 6, 2026
* data-flow F1+F2 P5.1: AzureHubRepository skeleton + composition root
Adds AzureHubRepository implementing HubRepository from @variscout/core/persistence.
Dispatch short-circuits HUB_PERSIST_SNAPSHOT to saveProcessHubToIndexedDB (bootstrap);
all other action kinds throw explicit P5.2/P5.3 not-yet-implemented error. Read APIs
wire against existing Azure Dexie tables (processHubs, evidenceSources, evidenceSnapshots,
evidenceSourceCursors); F3-pending entities (investigations, findings, questions,
causalLinks, suspectedCauses) are stubbed. Module-scoped azureHubRepository singleton
mirrors PWA composition-root pattern. apps/azure/CLAUDE.md updated with
persistence-boundary invariant documenting R12+R13 exceptions.
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P5.1 review fixes: hubs.list tombstone filter + multi-hub outcomes test + dispatch error test
- hubs.list() now filters h.deletedAt === null; hubs.get stays unscoped (comment added)
- outcomes.get has O(n hubs) comment noting F3 normalization will remove the scan
- AzureHubRepository.read.test: hubs.list tombstone test + multi-hub outcomes.get test
- AzureHubRepository.test: HUB_PERSIST_SNAPSHOT error propagation test
- Fix 4: corrected misleading fake-indexeddb comment ("kept at line 1" → "must be the first import statement")
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P5.2: Azure cascade helper with Dexie transaction
Adds cascadeArchiveDescendants(parentKind, parentId, archivedAt) in
apps/azure/src/persistence/cascadeArchive.ts. Uses transitiveCascade()
from @variscout/core/persistence; wraps all Dexie writes in a single
'rw' transaction covering evidenceSnapshots, evidenceSources,
evidenceSourceCursors, and processHubs. Per-kind stubs with F3 comments
for investigation/finding/question/causalLink/suspectedCause/rowProvenance/
canvasState (no Azure tables today). Outcome cascade explicitly deferred
to P5.3 per-action handler (hub blob mutation). Evidence source cursor
hub-range query uses compound PK [hubId+sourceId] between(). 10 new tests
covering: evidenceSource→cursor cascade, hub multi-table cascade,
investigation no-op, rollback, idempotency, empty-descendants. Barrel
updated to re-export the helper.
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P5.2 review fixes: cascade JSDoc + rollback comment + fan-out test
- Fix 1: add ## Atomic call pattern (P5.3+) section to cascadeArchiveDescendants
JSDoc prescribing the outer-transaction wrapping pattern P5.3 must use to
keep parent-row update + descendant cascade atomic; explains Dexie 4 zone reuse.
- Fix 2: replace hedge comment in rollback test with confident pass-confirmation
statement — rollback path verified green at P5.2 baseline (1098 tests).
- Fix 3: add cursor2 for src2 in the hub multi-table cascade test to cover the
fan-out path (hub → both evidenceSource cursors via compound key-range scan).
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P5.3: Azure per-action handlers + applyAction module
Implements all HubAction handlers in a new applyAction.ts module (Option B):
- Hub-blob mutations (HUB_UPDATE_GOAL, HUB_UPDATE_PRIMARY_SCOPE_DIMENSIONS,
OUTCOME_ADD, OUTCOME_UPDATE, OUTCOME_ARCHIVE) — read-modify-write via
saveProcessHubToIndexedDB.
- Direct Dexie table writes (EVIDENCE_ADD_SNAPSHOT, EVIDENCE_ARCHIVE_SNAPSHOT,
EVIDENCE_SOURCE_ADD, EVIDENCE_SOURCE_UPDATE_CURSOR, EVIDENCE_SOURCE_REMOVE).
- EVIDENCE_SOURCE_REMOVE atomically wraps cascade + parent soft-delete in a
single db.transaction; outer table list is a superset of cascadeArchiveDescendants'
internal transaction to satisfy Dexie zone propagation.
- Session-only kinds (INVESTIGATION_*, FINDING_*, QUESTION_*, CAUSAL_LINK_*,
SUSPECTED_CAUSE_*) — documented no-ops with F3 comment.
- Canvas kinds — documented no-ops with HUB_PERSIST_SNAPSHOT-flow comment.
- TypeScript exhaustiveness via assertNever() ensures compile-time completeness.
AzureHubRepository.dispatch now delegates to applyAction after the
HUB_PERSIST_SNAPSHOT short-circuit.
Adds 73 new tests (1098 → 1171). Adds sustainment-deferral note to
apps/azure/CLAUDE.md Invariants section.
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P5.3 review fixes: parent-update rollback test + @internal barrel + idempotency notes
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P6: Azure call-site migration to dispatch (storage facade + cursor sync)
Migration A: replace saveProcessHubToIndexedDB, saveEvidenceSourceToIndexedDB,
saveEvidenceSnapshotToIndexedDB in the three StorageProvider save methods with
azureHubRepository.dispatch(HUB_PERSIST_SNAPSHOT | EVIDENCE_SOURCE_ADD |
EVIDENCE_ADD_SNAPSHOT). Cloud-sync blocks unchanged. List/cache paths (lines 548,
595, 642) intentionally kept on direct localDb helpers (read-sync only).
Bootstrap save in listProcessHubs (line 548) left on direct call — it syncs cloud
→ local cache, not a hub-write dispatch path.
Migration B: replace db.evidenceSourceCursors.{get,put} in useEvidenceSourceSync
with azureHubRepository.evidenceSources.getCursor (Option B-1) and
dispatch(EVIDENCE_SOURCE_UPDATE_CURSOR). No direct Dexie usage remains in the file.
Tests: storage.test.ts — added db.processHubs mock for ensureDefaultProcessHubInIndexedDB;
added persistence mock for azureHubRepository.dispatch; added saveProcessHub /
saveEvidenceSource / saveEvidenceSnapshot tests asserting dispatch called with correct
action shapes. useEvidenceSourceSync.test.ts — replaced db.evidenceSourceCursors mock
with azureHubRepository mock; updated markSeen test to assert dispatch call shape.
Tests: 1173 → 1178 ✓ (5 new). Build: clean.
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P6 review fix: tighten cloud-sync test mocks for evidence paths
Add hoisted mocks + vi.mock factory entries for all six evidence-blob
functions (listBlobEvidenceSnapshots, saveBlobEvidenceSnapshot,
updateBlobEvidenceSnapshots, listBlobEvidenceSources,
saveBlobEvidenceSource, updateBlobEvidenceSources). Previously the
saveEvidenceSnapshotToCloud path called undefined blob functions which
were silently swallowed, letting the cloud-sync assertion pass without
the blob calls ever firing. Tighten the existing snapshot online test
to assert exact dispatch shape (provenance: []) + blob function calls.
Add parallel online + cloud-sync test for saveEvidenceSource covering
dispatch shape and blob function calls. Tests: 1178→1179.
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P7: ESLint dexie/db guard + dispatch-only docs
- P7.1: Update packages/stores, apps/pwa, apps/azure CLAUDE.md files to
explicitly state the dispatch-only invariant (domain stores never import
dexie directly; access via HubRepository / pwaHubRepository.dispatch /
azureHubRepository.dispatch). Update "will add" references to past tense
now that P7.2 delivers the rule. Enumerate R12+R13 exceptions clearly in
each file.
- P7.2: Add ESLint no-restricted-imports rule in eslint.config.js blocking
dexie package imports and **/db/schema glob imports outside the documented
allow-list (persistence/, db/, wallLayoutStore.ts, azure services/storage
+ localDb + cloudSync + lib/persistence, test files). Smoke-tested: rule
fires on ProcessHubEvidencePanel.tsx with clear actionable message; passes
cleanly against current codebase (0 errors).
- P7.3: Confirmed pnpm lint is already a step in pr-ready-check.sh (no
script change needed). pr-ready-check passes end-to-end.
Co-Authored-By: ruflo <ruv@ruv.net>
* data-flow F1+F2 P8.3 closeout: doc polish + decision-log + spec delivered
- Minor #1: add bootstrap cache-fill comment above storage.ts:548 for-loop
- Minor #2: R12 self-document comment on wallLayoutStore Dexie import
- Minor #3: fix plan frontmatter category (implementation-plan/audit → implementation)
- decision-log: F1+F2 closeout entry (what/why/deltas) + session backlog row
- spec: 2026-05-06-data-flow-foundation-design.md status active → delivered
Co-Authored-By: ruflo <ruv@ruv.net>
---------
Co-authored-by: ruflo <ruv@ruv.net>
jukka-matti
added a commit
that referenced
this pull request
May 8, 2026
8-task TDD-shaped plan executing the spec at docs/superpowers/specs/2026-05-07-canvas-hypothesis-arrow-drawing-design.md. Tasks (bottom-up): 1. useHypothesisDrawTool state machine + endpoint resolver (hooks) 2. activeCanvasTool field + auto-enable hypotheses overlay (hooks) 3. HypothesisDrawToolButton chrome component (ui) 4. HypothesisDraftPopover form (ui, hand-rolled positioning) 5. StepNodeMarker promoted-hypothesis pip (ui, Lucide Flag) 6. CanvasStepCard updates: replace badge, data-arrow-endpoint, gate click during draw 7. Canvas chrome integration: button + pointer handlers + rubber-band SVG + state-machine wiring through CanvasWorkspace 8. CanvasStepOverlay Remove button + investigations.md resolution + manual chrome walk steps Plan-time deferrals locked: column resolution via card.metricColumn (Risk #1); return-to-select after Save (Risk #2); hand-rolled positioning, no new dep (Risk #5); Lucide Flag with status colors (Risk #6). Subagent-driven dispatch per master plan D6 + feedback_subagent_driven_default. Sonnet for >=70% of dispatches; Opus for final review only. Master plan section 4 8d entry updated with forward link to this plan. Co-Authored-By: ruflo <ruv@ruv.net>
This was referenced May 9, 2026
jukka-matti
added a commit
that referenced
this pull request
May 14, 2026
…leanup workstream) (#166) * fix(8f-followup): delete legacy variscout-wall-layout Dexie DB on init Closes 8f followup HIGH #3 — pre-8f users carried an orphan IndexedDB forever after the wallLayoutStore → canvasViewportStore shape change. Mirror PwaHubRepository's legacy-DB cleanup pattern. Tightens the existing test that lied about asserting deletion. Co-Authored-By: ruflo <ruv@ruv.net> * refactor(8f-followup): migrate canvas UI strings to typed message catalogs Closes 8f followup HIGH #5 — 47 hardcoded English strings across SystemLevelView, CanvasLensPicker, MobileLevelPicker, NoFocalStepPrompt, AuthorL3View, LocalMechanismView now route through MessageCatalog. CANVAS_LENS_REGISTRY labels/descriptions translated at render time in CanvasLensPicker via LENS_LABEL_KEY / LENS_DESC_KEY maps; the hooks registry keeps English for non-UI consumers. Non-English locales receive English placeholders pending a translation pass (TODO(i18n) comment). Co-Authored-By: ruflo <ruv@ruv.net> * fix(8f-followup): migrate Canvas empty-state to message catalog Closes the i18n reviewer's flagged follow-up: Canvas/index.tsx:1042 rendered the lens registry's English label and a local CANVAS_LEVEL_LABELS map as user-facing copy, bypassing i18n. - Export LENS_LABEL_KEY from CanvasLensPicker for cross-component reuse - Reuse canvas.mobile.{system,process,step} for level labels - Add canvas.lensPicker.invalidAtLevel parameterized key (32 locales, English placeholder elsewhere per the catalog's translation pass plan) - Drop CANVAS_LEVEL_LABELS + remove now-unused CANVAS_LENS_REGISTRY import Co-Authored-By: ruflo <ruv@ruv.net> * test(8f-followup): cover CanvasLensPicker lens × level predicate Closes 8f followup LOW #20 — load-bearing CanvasLensPicker had no dedicated test. 18 enabled/disabled cells (3 levels × 6 lenses) + click-dispatch + aria-label assertions. Co-Authored-By: ruflo <ruv@ruv.net> * docs(8f-followup): refresh stale wallLayoutStore references in store comments Closes 8f followup LOW #18 — viewStore.ts:140 + preferencesStore.ts:178 still mentioned wallLayoutStore in doc-strings after the PR1 rename to canvasViewportStore. Co-Authored-By: ruflo <ruv@ruv.net> * docs(8f-followup): fix plan frontmatter category to allowed enum value * refactor(8f-followup): extract getStepColumnAssignments to @variscout/core/frame Closes 8f followup HIGH #2 — AuthorL3View's private focalStepColumns helper duplicates business logic that should live in core/frame. The helper now lives at packages/core/src/frame/stepColumns.ts with 6 unit tests; AuthorL3View imports it. Per ADR-074 amendment + ADR-081: Canvas embeds owner-surface computation, doesn't re-derive. Co-Authored-By: ruflo <ruv@ruv.net> * fix(8f-followup): tie L1 specLimits to outcome's own measureSpecs entry Closes 8f followup MEDIUM #8 — SystemLevelView trusted a flat specLimits prop without ADR-073-anchored contract. Now accepts measureSpecs keyed by column and derives from measureSpecs[map.ctsColumn] internally; old prop renamed to specLimitsOverride (advisory/debug only, deprecated). Canvas passes { [ctsColumn]: { usl, lsl, target, cpkTarget } } as measureSpecs. Two regression tests assert the leak scenario: wrong-column measureSpecs key produces '--' Cpk, not a silently wrong value. Co-Authored-By: ruflo <ruv@ruv.net> * refactor(8f-followup): replace LocalMechanismView's focalStepColumns duplicate Same ADR-074 amendment violation that PR2 fixed in AuthorL3View. The private helper now delegates to getStepColumnAssignments from @variscout/core/frame (introduced in the prior commit), flattening the structured result into the string[] this view needs. Co-Authored-By: ruflo <ruv@ruv.net> * docs(8f-followup): resolve lens × level matrix gap via spec amend Closes 8f followup HIGH #4 via AMEND path (not expand). Git blame shows both `performance` and `yamazumi` lenses were introduced with `enabled: false` AND registry descriptions explicitly labeling them as "Future ... lens" — intentional V2 placeholders, not bugs. Spec §10 was over-promised at original ship. - Spec §10 matrix amended: 6 cells marked `(V2 — deferred; lens not enabled in V1)` instead of pretending they ship enabled - V2 expansion path documented inline - investigations.md entry marked RESOLVED 2026-05-13 with rationale Co-Authored-By: ruflo <ruv@ruv.net> * feat(8f-followup): replace setViewportLevel throw with warn + no-op (4.4) l3 without focalStepId now emits console.warn and returns the viewport unchanged instead of throwing. fitToContent guards the same path. Updated test asserts warn was called and state did not change. Co-Authored-By: ruflo <ruv@ruv.net> * refactor(8f-followup): co-locate level math constants in core/canvas/viewport (4.7) Move FIT_TO_CONTENT_ZOOM_BY_LEVEL from canvasViewportStore into @variscout/core/canvas/viewport.ts and re-export it through the barrel. Add LOD_SNAP_BOUNDARIES (L2_OVERVIEW_LOW=0.5, L2_DETAIL_HIGH=1.8) for the upcoming snap-to-LOD feature. Single source of truth for level math. Co-Authored-By: ruflo <ruv@ruv.net> * feat(8f-followup): enforce 6px click-vs-drag deadband via clickDistance(6) (4.3) Adds .clickDistance(6) to the d3-zoom behavior in useCanvasViewportInput. Pointer moves ≤5px are treated as clicks; ≥6px as drags. Matches spec §6.3. Co-Authored-By: ruflo <ruv@ruv.net> * chore(8f-followup): delete dead worldToWallSvg + document CanvasViewport (4.5/4.6) worldToWallSvg was an identity function with no callers outside its own test; deleted function and test. CanvasViewport is used in Canvas/index.tsx — added JSDoc comment explaining its role so the seam is documented. Co-Authored-By: ruflo <ruv@ruv.net> * feat(8f-followup): snap-to-LOD on wheel-stop via d3-zoom end handler (4.2) Adds a 'end' listener to the zoom behavior. When the user releases the wheel with zoom in [0.3, 0.5) or [1.8, 2.0), the viewport eases back to 0.5 or 1.8 respectively over 150ms. Exports snapTarget() for unit testing. LOD_SNAP_BOUNDARIES lives in @variscout/core/canvas/viewport alongside LOD_THRESHOLDS. Co-Authored-By: ruflo <ruv@ruv.net> * feat(8f-followup): real LOD cross-fade + d3-transition snap (4.1/4.2 final) LODSwitcher now renders both outgoing and incoming renderers in stacked absolute divs during a 150ms window, then unmounts the outgoing. Uses useState+useEffect+setTimeout — no external animation library needed. useCanvasViewportInput snap-to-LOD uses d3-transition via selection.transition().duration(150).call(zoomBehavior.transform, ...). Adds d3-transition + @types/d3-transition to @variscout/hooks deps. Tests: 4 LODSwitcher tests assert dual-render during transition and single-render after 150ms via fake timers. Co-Authored-By: ruflo <ruv@ruv.net> * feat(8f-followup): per-Hub canvas viewport blob helpers in blobClient Closes 8f followup HIGH #1 part 1/2 — adds loadBlobCanvasViewport + saveBlobCanvasViewport mirroring the updateBlobEvidenceSnapshots ETag-conditional pattern. Per ADR-081 §2 (Azure = IndexedDB + Blob sync with ETag per ADR-079) and ADR-079. Also adds getLocalViewportUpdatedAt to @variscout/stores so the Azure lifecycle hook can compare timestamps without reading Dexie directly. Co-Authored-By: ruflo <ruv@ruv.net> * feat(8f-followup): wire Azure canvas viewport lifecycle to Blob sync Closes 8f followup HIGH #1 part 2/2 — useCanvasViewportLifecycle (Azure) now rehydrates from Blob after Dexie cache, debounced-persists to both Dexie and Blob with ETag, and treats precondition-failed as last-write- wins per spec §11 with App Insights telemetry on the conflict path. Co-Authored-By: ruflo <ruv@ruv.net> * feat(8f-followup): expose 4 remaining response-path CTAs at L3 column granularity Closes 8f followup MEDIUM #9 — LocalMechanismView previously only exposed Quick Action. Spec §5.3.a lists 5 CTAs at column-mechanism granularity (Quick Action / Focused Investigation / IP / Sustainment / Handoff). Threaded the 4 step-level callbacks already on CanvasProps through to LocalMechanismView; per-column button row added with new i18n keys (8 new MessageCatalog keys across 32 locale files). Parent callbacks are step-only; column is visible only within the card row. Co-Authored-By: ruflo <ruv@ruv.net> * feat(8f-followup): mobile L3 without focalStepId navigates to step-list Closes 8f followup MEDIUM #10 — MobileLevelPicker previously called setLevel('l2') before setZoom(2.5) as an explicit l2 redirect comment implied. The final committed state was already l3 (setZoom(2.5) fires inferLevel→l3), but the intent was undocumented. Updated comment to clarify the atomicity: both calls run synchronously before React re-renders, so the final state is l3 with no focalStepId, and canvas renders NoFocalStepPrompt (the step-list surface) per spec §7. Co-Authored-By: ruflo <ruv@ruv.net> * perf(8f-followup): selector-scope canvasViewport subscribe in d3-zoom hook Closes 8f followup MEDIUM #11 — useCanvasViewportInput previously subscribed to the whole canvasViewportStore; every unrelated mutation (setRailOpen, setViewMode, openChartCluster, etc.) fired syncElementToStoreViewport (which has its own diff-check guard, but still did needless work). Now tracks prevViewportRef and short-circuits on reference equality of state.viewports[hubId] — sync is skipped entirely when the hub's viewport slice hasn't changed. Test added: setRailOpen → d3 element __zoom unchanged. Co-Authored-By: ruflo <ruv@ruv.net> * chore(8f-followup): rename canvasViewport STORE_LAYER to annotation-per-hub Closes 8f followup LOW #15 — canvasViewportStore state is keyed by hubId not projectId since the 8f shape change. The annotation-per-project label was technically truthful (per-project umbrella, hub-keyed inside) but invited confusion. Renamed to annotation-per-hub; layerBoundary test + packages/stores/CLAUDE.md table updated. 'annotation-per-project' is now in the reserved/unused set; 'annotation-per-hub' is live. Co-Authored-By: ruflo <ruv@ruv.net> * docs(8f-followup): mark 19 of 20 findings RESOLVED on followup branch --------- Co-authored-by: ruflo <ruv@ruv.net>
7 tasks
jukka-matti
added a commit
that referenced
this pull request
May 16, 2026
…lar) Project tab is active-IP-centric — shows ONE project's lifecycle (amendment spec §Decision summary #2). Rename key across all 32 locale files + MessageCatalog type + PWA nav consumer; update values to singular form for inflecting languages (en/fi/de/es/fr/it/pt/nl/pl/tr/sv/da/nb/cs/hu/ro/uk/sk/hr/bg/el/hi/ar/he) and native equivalents for non-inflecting languages (ja/ko/zh-Hans/zh-Hant/th/vi/id/ms). TDD: failing tests written first, confirmed red, then green.
jukka-matti
added a commit
that referenced
this pull request
May 16, 2026
* docs(plans): bite-sized sub-plan for PR-WV1-2 Improve workspace migration
* refactor(ui): route IPDetailPage + Charter ACL through canAccess
* docs(ui): explain CharterOverview empty-members escape
* feat(core): add migrateImprovementProjectMetadata + rename Handoff stage to Improve
Introduces migrateImprovementProjectMetadata (TDD, 4 tests) that
idempotently folds legacy team[] → wedge members[] at hydration time.
Exports from the improvementProject barrel.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(ui,apps): wire stage rename + metadata migration into hydration paths
Renames StageName / StageStateMap from handoff→improve; stage order is
now charter→approach→improve→sustainment. Updates IPDetailPage body render,
defaultActiveStage, and all affected tests. Wires
migrateImprovementProjectMetadata into Azure (hydrateHub) and PWA
(joinHub) at the lowest Dexie read seam.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(ui): add ImproveStage with canAccess-gated ActionItem tracker
Implements Task 2 of PR-WV1-2: creates ImproveStage component backed by
ActionItem list with edit-improve ACL gating, empty-state, add form, and
owner/due-date display; routes IPDetailPage 'improve' stage to ImproveStage
replacing the Handoff placeholder; adds onActionAdd/Update/Remove optional
props (no-op defaults) to IPDetailPageProps.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(ui): add ImproveStageAdvanced mounting PDCA primitives
Composes ImprovementContextPanel, IdeaGroupCard (per group), BrainstormModal
(modal-trigger owned by wrapper), PrioritizationMatrix, and WhatIfExplorer
into the Advanced workspace four-region layout.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(ui): add Advanced toggle on ImproveStage
Progressive disclosure per wedge spec §3.3 — simple action tracker by
default; Advanced button swaps in ImproveStageAdvanced (PDCA workbench).
Toggle visible to all roles; edit gating remains inside each view.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(ui): fold Handoff close-logic into Sustainment closure
Wedge V1: the Handoff stage is retired; its close-project checklist
(control plan, training, cadence, process-owner acknowledgment + nudge)
is absorbed into SustainmentOverview as an optional closureInputs panel.
SustainmentSections gains a controlHandoff section in its sections view.
HandoffOverview and HandoffSections source files deleted; HandoffOverview
tests ported into SustainmentOverview test suite (4 new closure-panel
assertions). HandoffChecklistInputs type replaced with
SustainmentClosureInputs exported from SustainmentOverview.
IPDetailPage, both ProjectsTabViews (PWA + Azure), and App.tsx / Editor.tsx
callers updated: handoffInputs → closureInputs, onOpenLegacyHandoff removed.
Pre-existing whatIfMode:'basic' tsc error in ImproveStage absorbed and fixed
(fallback 'basic' → 'standard').
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* docs(specs): wedge V1 amendment — Improve top-level tab + Project singular
* docs(plans): bite-sized amendment plan for Improve-tab restoration
* refactor(ui): trim StageName to 3 values — drop 'improve' stage per amendment
Removes 'improve' from StageName, StageStateMap, and deriveStageState per the
2026-05-16 wedge amendment (Improve moves to top-level tab; Project detail shows
Charter / Approach / Sustainment only). Adds 'upcoming' to StageState union.
* refactor(ui): remove ImproveStage routing from IPDetailPage per amendment
Drop the 'improve' stage case, its render block, and the three onAction*
props from IPDetailPage — Improve is now a top-level tab, not a project
detail stage. Also remove the dead stages.improve branch from
defaultActiveStage and trim the matching test assertions.
* feat(ui): add NoActiveProjectGuidance empty state for Improve tab
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(ui): move ImproveStage to Improve/; retire ImproveStageAdvanced; reuse ImprovementWorkspaceBase as Advanced view
Advanced toggle now renders ImprovementWorkspaceBase (production PDCA workbench)
instead of the skeleton ImproveStageAdvanced. Test mocks ImprovementWorkspaceBase
to keep the unit hermetic; runtime integration verified via --chrome browser walk.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(ui): add ImproveTabRoot orchestration component
Routes between NoActiveProjectGuidance empty state and ImproveStage
based on activeIP. Adds Improve/ barrel + re-exports from @variscout/ui.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(apps): wire ImproveTabRoot into PWA + Azure Improve tab
PWA: ImprovementView gains activeIP + onGoHome props; when activeIP is
threaded from App.tsx, renders ImproveTabRoot (NoActiveProjectGuidance
when null, action tracker when set). Fallback to pre-wedge ideas view
when activeIP prop not provided for backward compatibility.
Azure: Editor.tsx Improve section adds activeIP === null guard that
renders ImproveTabRoot empty state; active-IP case falls through to
existing ImprovementWorkspaceBase ideas workspace unchanged.
Action callbacks are console.warn stubs pending PR-WV1-3 persistence work.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(apps): route Azure Improve tab through ImproveTabRoot + tighten PWA prop contract
Azure Editor was gating <ImproveTabRoot> behind activeIP===null, letting active-IP
users fall through to the legacy <ImprovementWorkspaceBase> ideas view. Removed the
ternary — both branches are now a single unconditional <ImproveTabRoot> that handles
empty and active-IP states internally (mirrors PWA behaviour). Removed now-dead
imports and variables (ImprovementWorkspaceBase, ImprovementContextPanel,
WhatIfExplorerPage, PrioritizationMatrix, TrackView, DEFAULT_PRESETS, matrix axis
state, activeImprovementView/highlightedIdeaId store reads, scopedImprovementQuestions,
and unused orchestration destructure fields).
PWA ImprovementView: made activeIP, actions, onGoHome required (no optional shims);
deleted the pre-wedge fallback branch (ideas workspace + empty-ideas guard). App.tsx
call site updated to pass actions={[]} explicitly, drop the seven props that no longer
exist on the component, and remove the now-unused useImprovementOrchestration hook call.
Per feedback_no_backcompat_clean_architecture.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(i18n): rename workspace.projects to workspace.project (singular)
Project tab is active-IP-centric — shows ONE project's lifecycle (amendment spec §Decision summary #2). Rename key across all 32 locale files + MessageCatalog type + PWA nav consumer; update values to singular form for inflecting languages (en/fi/de/es/fr/it/pt/nl/pl/tr/sv/da/nb/cs/hu/ro/uk/sk/hr/bg/el/hi/ar/he) and native equivalents for non-inflecting languages (ja/ko/zh-Hans/zh-Hant/th/vi/id/ms). TDD: failing tests written first, confirmed red, then green.
* docs(wedge): log PR-WV1-2 delivery + Improve-tab amendment
Records the Improve-tab-restore amendment under the 2026-05-16 wedge
entry: Improve restored as top-level verb tab, Project singular rename,
StageName trimmed to 3, ImproveTabRoot wired in both apps, deferred
items (b) + (d) from PR-WV1-1 closed, (a) + (c) still owed to
PR-WV1-3 / PR-WV1-5.
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
jukka-matti
added a commit
that referenced
this pull request
May 17, 2026
…log format + classify agent manifests Audit gaps surfaced after yesterday's discipline shipment: 1. "One canonical home per concept" was buried as an anti-pattern; now stated affirmatively as Principle #2 2. Decision-log entry format was informal; now codified with edit-type vocabulary for Play 2c toolbox parsing 3. CLAUDE.md/AGENTS.md/llms.txt weren't classified in the doc-types table Both strategy spec §2.7 and doc-discipline.md updated in parallel: Core principles (in priority order): 1. Doc-type discipline 2. One canonical home per concept (Play 4 enforces) 3. Reader-first banners 4. Decision-log as temporal index Wedge-amendment incident reframed as a simultaneous violation of principles 1+2+3, making clear why all four need mechanical enforcement (Play 2b). Doc types extended in the at-a-glance list: investigations + memory + agent manifests added (was 5 types covered, now 8). Agent manifests get an explicit row in the doc-discipline.md table noting they're the "first read" surface every agent dispatch loads — keep tight, CLAUDE.md size budget enforced. Decision-log entry standard format: - YYYY-MM-DD — <title ≤80c>. <edit-type>: <doc>#<section> [supersedes <prior>]. Why: <one-sentence>. Commit: <sha>. PR: #N. Related: [[<id>]]. With edit-type vocabulary (spec edit / ADR amendment / new ADR / supersession / archived / new spec) for machine-parseable categorization by Play 2c toolbox. Lists what NOT to put in entries (no rationale paragraphs, no restated spec content, no duplicated info — one decision = one entry). Strategy spec banner updated to reflect today's second material edit. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
jukka-matti
added a commit
that referenced
this pull request
May 17, 2026
… latest main) (#199) * docs: foundational artifacts for docs-strategy-2026 (memo + spec + ADR-083) Create three foundation docs that the rest of Play 1 rests on: - 2026-05-16-docs-strategy-memo.md: 1-page CTO memo (§8 of plan) - 2026-05-16-docs-strategy-design.md: full strategy spec (§§1-7, 10) - adr-083-eight-purpose-doc-taxonomy.md: ADR for the schema change ADR-083 captures the 8-purpose × 4-tier two-axis schema, STATUS alias map (22→4), AUDIENCE alias map (14→3), CATEGORY→topic-tags replacement, and the migration approach for Play 1. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(scripts): collapse frontmatter schema (22→4 STATUS, free-form topic, 3 AUDIENCE) + add purpose/tier + drift fields Per ADR-083: two-axis schema (8 purposes × 4 tiers) replaces the over-engineered 22-STATUS × 18-CATEGORY × 14-AUDIENCE triplet. scripts/docs-frontmatter-schema.mjs: - Add PURPOSE enum (8 values) and TIER enum (4 values) - Collapse STATUS from 22 → 4 canonical values - Export STATUS_ALIAS_MAP (11 old→new mappings) for transitional validation - Export AUDIENCE_ALIAS_MAP (14 old→new mappings) - Remove CATEGORY enum; topic tags are now free-form - Collapse AUDIENCE from 14 → 3 canonical values (human/agent/both) - Add new optional fields: purpose, tier, last-verified, verified-against-commit, supersedes, related - Update classify() for new folder paths (Play 1 prep) scripts/check-doc-frontmatter.mjs: - Warn (not fail) on aliased STATUS/AUDIENCE values - Hard-fail on unknown PURPOSE/TIER when present - Add --report mode: counts docs missing purpose + tier scripts/docs-frontmatter-fix.mjs: - Apply STATUS_ALIAS_MAP and AUDIENCE_ALIAS_MAP - Handle inline array audience form (audience: [analyst, engineer]) - Backfill purpose + tier via path heuristics (all 521 docs covered) Applied to all 521 docs: 0 hard violations, 0 transitional warnings. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(agent-context): add 3 Tier 1 supporting docs (quickstart + package-router + store-state-glossary) New docs at docs/agent-context/ (TEMP location; will move to docs/living/agent-context/ in Play 1b): - onboarding-quick-start.md: 5-minute agent orientation covering wedge V1 direction, hard invariants, where-to-look table, and common pitfalls (Math.random, hex colors, --no-verify, root-cause language) - package-router.md: 20-row routing table mapping work areas → primary CLAUDE.md + related CLAUDEs + ADRs + rules/invariants + suggested skills - store-state-glossary.md: full glossary of all 7 stores × 3 layers (Document/Annotation/View) per ADR-078 + F4, with top fields, persistence details, and hard rules per store All 3 docs use the new two-axis schema (purpose, tier, audience, topic, last-verified) introduced in the docs-strategy-2026 plan. Verified against packages/stores/src/*.ts at commit 5ee58dc. pnpm docs:check passes (527 docs validated, no issues). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(skills): add 3 Tier 1 agent-context skills (quickstart + package-router + store-state-glossary) Closes the 4%-of-living agent-context gap identified in plan §7.1. Three skills for fresh subagent dispatch orientation (Play 3 Tier 1): - agent-context-quickstart: loads docs/agent-context/onboarding-quick-start.md; description triggers on any "start of session", "new subagent", or "need context" phrasing; covers wedge V1 direction, hard invariants, where-to-look, and common pitfalls - package-router: routes work areas (canvas, stats, charts, stores, i18n, azure, pwa, ui, hooks, coscout, response-paths, capability/yamazumi/defect/performance/process-flow modes, projects tab, investigation wall, evidence map, sample data) to primary CLAUDE.md + related CLAUDEs + ADRs + rules; description triggers on "which package", "where to look", "which CLAUDE.md" - store-state-glossary: covers all 7 stores × 3 layers per ADR-078 + F4; description triggers on "which store", "what state lives", "store state" phrasing; links to full doc and packages/stores/CLAUDE.md All 3 skills follow the SKILL.md spec (name ≤64 chars, description with what+when, body ≤200 lines). Skills use force-add to override .claude/.gitignore — standard pattern for project-scoped skills. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(.claude): rename .claude/rules/ → .claude/invariants/ + update path references Renames 6 constraint files (stats, testing, charts, coscout-prompts, i18n, azure-storage) from .claude/rules/ to .claude/invariants/ to clarify their purpose as policy-level invariants (not procedure routers). Updates all active references in CLAUDE.md, package-router.md, and two implementation docs to the new path. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(.claude): add INVARIANTS.md synthesis index Creates .claude/INVARIANTS.md as the single entry-point for all VariScout cross-cutting non-negotiables. Organises invariants into three tiers — hard (lint/test-enforced), soft (convention + reviewer), and topic-scoped (load on demand) — with canonical home and enforcement mechanism cited for each. Identifies two open enforcement gaps (Math.random and Tailwind @source) for follow-up in investigations.md. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs: replace prose invariant duplication with [[invariant]] links + log enforcement gaps Links root CLAUDE.md §Invariants and onboarding-quick-start.md §Hard Invariants to the new .claude/INVARIANTS.md index instead of restating the full invariant list. Updates agent-context-quickstart skill to explicitly load INVARIANTS.md in its How to Use procedure. Logs two open enforcement gaps (Math.random ban, Tailwind @source check) to docs/investigations.md as candidates for Play 7. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * feat(scripts): add docs:gen-arch — auto-generate dependency/exports/Tailwind architecture doc Adds scripts/docs/gen-arch.mjs that walks package.json files (dep graph), tsconfig.json paths (sub-path export map), and apps/*/src/index.css (@source directives) to emit docs/05-technical/architecture-generated.md. Wired as pnpm docs:gen-arch in root package.json. Pre-push hook regenerates and fails if checked-in version is stale, surfacing drift immediately. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * docs(system): add initial auto-generated architecture doc + cross-link from technical index Adds docs/05-technical/architecture-generated.md (initial committed version). Updates docs/05-technical/index.md to add inbound link so the doc is not an orphan (required by pnpm docs:check orphan check). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * docs(design): add canonical CoScout AX-design doc consolidating ADR-060/068/069 + prompt rule Creates docs/01-vision/coscout-ax-design.md as the single canonical AX-design surface for CoScout: persona, voice, tier-gated behaviors (table), five-pillar knowledge architecture, three-boundary safety, prompt engineering link, tool-calling, voice input, eval discipline, mode-aware methodology coaching, and related artifacts. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * docs: cross-link constituent CoScout ADRs + invariant + router to canonical AX-design - ADR-060, ADR-068, ADR-069: add related: [coscout-ax-design] to frontmatter + brief "See canonical AX design" note near each ADR title - .claude/invariants/coscout-prompts.md: add related: [coscout-ax-design] + see-also note at bottom - docs/agent-context/package-router.md: update CoScout row to point at coscout-ax-design.md as canonical entry point (was packages/core/CLAUDE.md) - .claude/INVARIANTS.md: update CoScout-prompts topic-scoped row to reference coscout-ax-design.md as the first-load for CoScout work Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * chore(prettier): exclude auto-generated arch doc + regenerate The Play 7 gen-arch script outputs compact markdown tables; lint-staged prettier reformatted them on commit, causing the pre-push staleness check (`git diff --exit-code` after regeneration) to fire on every push. Fix: add `.prettierignore` excluding the generated file so committed content matches the script's output verbatim. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(strategy): revise Play 2b to SSoT-by-doc-type discipline + ship convention doc Discipline gap surfaced by the 2026-05-17 wedge-amendment incident: a mid-flight design change to wedge spec was captured as `2026-05-16-improve-tab-amendment-design.md` (orphan side spec), invisible to a fresh session reading only the canonical wedge spec. Convention alone didn't prevent it; mechanical enforcement does. Strategy spec (edited in place, dogfooding the principle): - §1 Context: note the discipline-drift failure mode + why mechanical enforcement matters - §2.6 per-purpose table: design row → edit-in-place; remember row → append-only - §2.7 NEW: Doc Discipline (SSoT by doc type). Design specs edit-in-place; ADRs amendment-block-at-bottom; decision-log append-only; generated never-hand-edit; plan files ephemeral. Anti-pattern table + spec lifecycle states + edit-in-place mechanics + "decision-log as temporal index" framing. - Play 2b: revised from "addendum threads on all living docs" (wrong-shape — conflated design specs with ADRs) → "SSoT discipline by doc type" with validator forbidding `*-amendment-*.md` / `*-revision-*.md` / `*-update-*.md` filenames + lifecycle-state enforcement. Marked safe-parallel — ships now, no Play 1b dependency. - §4 migration: shipped/deferred status table reflecting current execution state - §10 risks: 3 new rows (discipline drift, validator false-positives, body staleness) Memo: aligned §What we're changing + §plays item 2 + §What we kill; added `last-verified`. NEW canonical: `docs/agent-context/doc-discipline.md` — focused reference (~150 lines) loaded by `agent-context-quickstart` skill on session start. Includes: - Doc-types + update-patterns table (7 types) - Anti-patterns table with right-pattern column - Spec lifecycle states (draft → active → delivered → superseded → archived) - Edit-in-place mechanics for design specs (5-step procedure) - Decision tree: "I need to change a canonical doc → which pattern?" - Origin story citing the wedge-amendment incident agent-context-quickstart SKILL.md: wires doc-discipline.md as step 4 ("If your task involves editing ANY canonical doc"). Added to Supporting Documents list. This commit dogfoods the principle: strategy spec edited in place, not as an `*-amendment-design.md` side file. Plan file at ~/.claude/plans/ updated separately with a banner pointing to the canonical spec as the SoT going forward. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * chore: regenerate arch doc Routine regen after staleness check (per Play 7 pre-push convention). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(discipline): add reader-first status banner principle (top of file) Question: should amendment/relevance info be at the top of files? Does wikilink structure preserve that? Answer: YES, top. Frontmatter is for tooling; banner is for readers (humans + agents). Wikilinks preserve graph structure but are passive at read time — banners are active. Industry convention (MDN, Stripe, IETF, Wikipedia, Tailscale): status at top, amendments at bottom (ADRs only). Added to doc-discipline.md §Reader-first banners: - Required/recommended matrix per condition (status: superseded, supersedes:, recent material edit, delivered, ADR with amendments) - 6 concrete banner templates (superseded, recent edit, active build, delivered, ADR amended, archived) - Origin story citing wedge-amendment incident - "Wikilinks don't substitute for banners" explanation - Banner + frontmatter both required (machine vs reader) - Play 2b validator extensions for banner enforcement (4 new rules) Added to strategy spec §2.7: - New subsection "Reader-first banners at the top" pointing to discipline doc - Edit-in-place mechanics step 3 now includes "add status banner" - Play 2b validator scope extended in mechanics section Dogfooded: strategy spec itself now carries a banner at top noting active build via PR #184 + last material edit 2026-05-17. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(discipline): elevate 4 affirmative principles + codify decision-log format + classify agent manifests Audit gaps surfaced after yesterday's discipline shipment: 1. "One canonical home per concept" was buried as an anti-pattern; now stated affirmatively as Principle #2 2. Decision-log entry format was informal; now codified with edit-type vocabulary for Play 2c toolbox parsing 3. CLAUDE.md/AGENTS.md/llms.txt weren't classified in the doc-types table Both strategy spec §2.7 and doc-discipline.md updated in parallel: Core principles (in priority order): 1. Doc-type discipline 2. One canonical home per concept (Play 4 enforces) 3. Reader-first banners 4. Decision-log as temporal index Wedge-amendment incident reframed as a simultaneous violation of principles 1+2+3, making clear why all four need mechanical enforcement (Play 2b). Doc types extended in the at-a-glance list: investigations + memory + agent manifests added (was 5 types covered, now 8). Agent manifests get an explicit row in the doc-discipline.md table noting they're the "first read" surface every agent dispatch loads — keep tight, CLAUDE.md size budget enforced. Decision-log entry standard format: - YYYY-MM-DD — <title ≤80c>. <edit-type>: <doc>#<section> [supersedes <prior>]. Why: <one-sentence>. Commit: <sha>. PR: #N. Related: [[<id>]]. With edit-type vocabulary (spec edit / ADR amendment / new ADR / supersession / archived / new spec) for machine-parseable categorization by Play 2c toolbox. Lists what NOT to put in entries (no rationale paragraphs, no restated spec content, no duplicated info — one decision = one entry). Strategy spec banner updated to reflect today's second material edit. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(agent-context): align with current main reality — 7-tab nav, €120, 3 personas, Math.random ESLint enforced Stale-reference fix pass after cherry-picking PR #184's infrastructure onto fresh branch off latest main. Aligns agent-context surface with what shipped today: - `docs/agent-context/onboarding-quick-start.md`: - €99 → €120; 6-tab → 7-tab nav - "Improve = stage inside Projects" → "Improve = top-level verb tab" (per 2026-05-16 Improve-tab amendment); Project stages collapsed to Charter→Approach→Sustainment - Add 3 personas inside Project: Lead (full edit + manages membership) / Member (full edit) / Sponsor (Report-only at V1, signoff out-of-band). Every subagent loading quickstart sees all 3 on first read. - Tier-gating "fully retired" (no more isPaidTier() checks) - VariScout Process named-future now points to `docs/01-vision/variscout-process/` - `.claude/skills/agent-context-quickstart/SKILL.md`: same updates (Key Facts section) - `docs/agent-context/package-router.md`: Projects-tab row updated — "Lead/Member/Sponsor roles" → "3 personas: Lead/Member/Sponsor"; stage list shortened; Improve as top-level noted; €120 pricing - Root `CLAUDE.md`: add "3 personas within each Project (Lead/Member/Sponsor)" to the strategic-direction line (the rest was already current — €120, 7-tab, Improve restored) - `.claude/INVARIANTS.md` §Enforcement Gaps: Math.random ban marked RESOLVED 2026-05-17 via PR #198 (Tier 1 Math.random retirement + ESLint guard shipped) Per user's terminology decision (2026-05-17): use "3 personas" framing consistently in agent-context docs (simpler than wedge spec's "1 persona + 3 ACL roles"). Specialist is the ICP description, not a 4th persona. Open question logged for follow-up: unify wedge spec terminology to match (out of scope for this PR). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(transcripts): promote docs-strategy-2026 brainstorm plan file to repo transcript Per the doc-discipline.md ephemeral-promotion rule: plan files in ~/.claude/plans/ are session-private by default; SELECTIVE promotion to docs/ephemeral/transcripts/ for LANDMARK sessions. The docs-strategy-2026 brainstorm qualifies — it produced the foundational strategy + ADR-083 + the SSoT-by-doc-type discipline. What this preserves in repo: - Full design journey (8 purposes × 4 tiers, audit findings from Agent 1 + Agent 2) - Mid-session pivots (wedge-conflict surfaced → split Play 1 into 1a safe + 1b deferred; original "addendum threads" Play 2b revised to SSoT discipline after the wedge-amendment incident) - §11 post-PR-184 recalibration (cherry-pick recovery plan + Play 1b/2a reorder + 3-personas terminology decision) Frontmatter added: purpose: remember, tier: ephemeral, status: archived, audience: human, topic: [docs-strategy, transcript, brainstorm]. Top banner converted from "plan-file pointing-at-canonical" to "archived transcript pointing-at-canonical" — the file IS now the canonical historical record. Inbound link added from strategy spec (avoids orphan check). Future readers see: spec for "what" (current intent); transcript for "why each decision was made". Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(schema): add 'named-future' to STATUS_ALIAS_MAP The user's PR #193 (VariScout Process named-future capture) created 8 docs at `docs/01-vision/variscout-process/` with `status: 'named-future'`. That value was valid in the old 22-value STATUS enum but missing from my collapsed 4-value canonical set. Maps to 'active' since named-future docs ARE actively maintained design artifacts for the future product — not draft (designed) or superseded (replaced). They describe what the named-future product WILL be. Resolves 8 hard-fail check-doc-frontmatter errors blocking push. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * docs(schema): named-future earns its own STATUS value (5 canonical, not 4) User feedback: aliasing 'named-future' to 'active' (commit 3a6928b) overstated the commitment level. VariScout Process docs at docs/01-vision/variscout-process/ are aspirational — contingent on V1 reaching 500+ customers. They're not in active design; they describe what's COMMITTED for a future product line. The 4-value collapse (draft/active/superseded/archived) was overzealous. The 22→4 reduction missed a meaningful distinct lifecycle state. Promoting 'named-future' to a canonical 5th status preserves the semantic intent: - draft: designing; not yet building - active: designed; build underway - **named-future: aspirational/conditional commitment** (not in active design) - superseded: replaced by another spec - archived: historical reference only Changes: - scripts/docs-frontmatter-schema.mjs: add 'named-future' to STATUS enum; remove from STATUS_ALIAS_MAP (no longer aliased) - docs/agent-context/doc-discipline.md: spec-lifecycle table extended; example cites VariScout Process docs; added note on 'delivered' being a soft state signalled by `delivered-by:` frontmatter (not a status: enum value) - docs/superpowers/specs/2026-05-16-docs-strategy-design.md: schema section + collapse summary + migration sequence updated (22→5 not 22→4) - docs/superpowers/specs/2026-05-16-docs-strategy-memo.md: same Resolves 8 hard-fail check-doc-frontmatter errors for variscout-process/ docs WITHOUT semantically mis-labeling them. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * chore: regenerate arch doc --------- Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bumps actions/upload-artifact from 4.6.2 to 7.0.0.
Release notes
Sourced from actions/upload-artifact's releases.
... (truncated)
Commits
bbbca2dSupport direct file uploads (#764)589182cUpgrade the module to ESM and bump dependencies (#762)47309c9Merge pull request #754 from actions/Link-/add-proxy-integration-tests02a8460Add proxy integration testb7c566aMerge pull request #745 from actions/upload-artifact-v6-releasee516bc8docs: correct description of Node.js 24 support in READMEddc45eddocs: update README to correct action name for Node.js 24 support615b319chore: release v6.0.0 for Node.js 24 support017748bMerge pull request #744 from actions/fix-storage-blob38d4c79chore: rebuild distDependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting
@dependabot rebase.Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebasewill rebase this PR@dependabot recreatewill recreate this PR, overwriting any edits that have been made to it@dependabot show <dependency name> ignore conditionswill show all of the ignore conditions of the specified dependency@dependabot ignore this major versionwill close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor versionwill close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependencywill close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)