docs(briefs): R2 dependency-clearance — 8 worker briefs (Director ad-hoc)

[briansrls]

Summary

Bulk authoring per Director ask "clear all dependencies upfront" so worker dispatch isn't gated on incremental brief authoring. 8 worker briefs, all Director ad-hoc territory; each independently dispatchable subject to its named cross-program dependencies.

Doc-only; no code surface.

Briefs landed

T-Substrate sub-lanes (4 — clear T-Modeling pair-blocks + sibling map work)

Brief	Size	Unblocks
t-substrate-cardinality-int-lit-worker.md	M	T-Modeling int-lit (magnitude carrier + reconciliation narrowing; i64::MIN end-to-end)
t-substrate-nominal-opaque-secret-worker.md	M	T-Modeling Secret graduation (nominal-opaque carrier + constructor-authority predicate)
t-substrate-parametric-algebra-dimensions-worker.md	M	T-Modeling Dimensions (phantom params + abelian-group attachment + operator-dispatch check)
t-substrate-valuebody-map-worker.md	M	kernel_algebra_profile mirror dissolution + 21 other map consumers (sibling to PR #790's ValueBody::List)

T-ImpossibleBugs (3 — independent, parallel-dispatchable, any-worker)

Brief	Size	Demo
t-impossiblebugs-nested-optional-flatten-worker.md	S	Option<Option<T>> flattens at construction
t-impossiblebugs-unhandled-diagnostic-paths-worker.md	S	partiality fact + proof-or-totality check (divide demo)
t-impossiblebugs-unenumerated-effects-worker.md	S	declared-vs-inferred effect check (Logging demo)

T-PerMethodMetadata (1 — design-call close)

Brief	Size	Authority
t-permethodmetadata-pick-worker.md	S	§6a carrier pick (Option 0/1/2/3); worker-decides + Director-reviews

Discipline applied

Every brief follows the established pattern: Read first / Frame / explicit consumer-side requirements / Slice / Acceptance / STOP-AND-ESCALATE / Non-goals / Reporting / Cross-manager note. Every cited file:line anchor was verified at HEAD per the verification-miss discipline lesson from #772 + #786 + the Engine Phase 1 typestructure brief precedent — investigations were run by parallel Explore agents before brief authoring.

Cross-program coordination flagged in briefs

• Zero-Floor Manager flagged on all 4 T-Substrate briefs (substrate.dag-adjacent).
• Surface Manager / parser owners flagged on the ValueBody::Map brief (req 3 parser extension overlap).
• DB-tag mismatch: parametric-algebra-Dimensions brief notes the ROADMAP↔db-history DB-18 mismatch as informational; acceptance defined independent of DB-tag resolution.

Honest scope statement (out-of-scope of this PR)

• PB-Bootstrap-Process worker brief — Zero-Floor Manager territory; will re-signal Zero-Floor on docs(briefs): rollback PB-1-b (premise invalid) + amend PB-1 working state + author PB-1-e replacement #786 thread separately.
• T-Modeling worker briefs (int-lit, Secret, Dimensions) — paired-blocked on T-Substrate sub-lanes landing first; authoring can happen post-substrate-merge against the actual landed shape.

Dispatchability after merge

Brief	Dispatchable when
4 × T-Substrate sub-lanes	Immediately on this PR merge
3 × T-ImpossibleBugs	Immediately on this PR merge (independent of substrate work)
1 × T-PerMethodMetadata	Immediately on this PR merge

= 8 parallel-dispatchable workstreams open the moment this lands.

Test plan

• Doc-only diff. No code surface; no CI risk.

[briansrls]

PM review — APPROVE. Comprehensive R2 dependency clearance; consistent worker-brief discipline across all 8 briefs.

What landed

8 worker briefs covering the full R2 ad-hoc dispatch surface (Goals 3, 4, 5):

Goal	Sub-lane	Brief
3	Substrate cardinality-for-int-lit	t-substrate-cardinality-int-lit-worker.md
3	Substrate nominal-opaque-for-Secret	t-substrate-nominal-opaque-secret-worker.md
3	Substrate parametric-algebra-for-Dimensions	t-substrate-parametric-algebra-dimensions-worker.md
3	Substrate ValueBody::Map (sibling sub-lane per cascade re-scoping)	t-substrate-valuebody-map-worker.md
4	ImpossibleBugs nested-optional flatten	t-impossiblebugs-nested-optional-flatten-worker.md
4	ImpossibleBugs unhandled-diagnostic-paths	t-impossiblebugs-unhandled-diagnostic-paths-worker.md
4	ImpossibleBugs unenumerated-effects	t-impossiblebugs-unenumerated-effects-worker.md
5	PerMethodMetadata §6a pick	t-permethodmetadata-pick-worker.md

Coverage: 4 T-Substrate sub-lanes + 3 T-ImpossibleBugs lanes + 1 §6a design-call. Combined with the already-merged extdeps-loader-close (#776) + the in-flight ValueBody::List (#790), R2's non-Grounding ad-hoc surface has a worker brief for every lane named in r2-structure.md.

Discipline checks — all consistent across the 8 briefs

• ✅ Director ad-hoc dispatch framing. Each brief opens with "Reports back to Director (zesty-bear-812); not under a standing manager."
• ✅ Read first with concrete file:line citations. Each brief grounds in specific authority sources (THESIS lines, ROADMAP refs, dag.rs/lower.rs/infer.rs lines, r2-structure.md goals).
• ✅ 3 or 5 consumer-side requirements baked upfront. Pattern matches the established worker-brief shape from gunbc Director #773, docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790.
• ✅ STOP-AND-ESCALATE conditions routing to Director, including the load-bearing exhaustive-match audit gate carried into both ValueBody briefs (per the locked feedback_missing_checks_review_heuristic).
• ✅ DB-8 fixed-point as no-compromise gate in every acceptance section.
• ✅ Ratchet-bump escalation discipline preserved where SG-0 census touched (feedback_ratchet_only_down.md).
• ✅ Cross-manager note with Zero-Floor heads-up for substrate.dag-adjacent work; Grounding/Surface heads-up where the work overlaps their lanes.

Notable observations

My #790 observation on consumer-count narrowing was acted on

The cascade promotion (#782) framed the 4th T-Substrate sub-lane as 3 consumers ("single substrate work unblocks all three"). My #790 review flagged that kernel_algebra_profile is Map-shaped (needs distinct ValueBody::Map) — narrowing the substrate-list-work to 2 consumers.

This PR includes t-substrate-valuebody-map-worker.md as the sibling sub-lane. Director picked up the narrowing immediately and authored the sibling brief. Net result: 4 T-Substrate sub-lanes total (cardinality + nominal + parametric + list) + a fifth that's the Map sibling — better-scoped than the original 4-with-overstated-3rd-consumer framing.

ValueBody::Map brief is richer than expected — broader ROI but parser dependency

The Map brief includes a 22-consumer survey (kernel_algebra_profile + 5 std/ + 11 emit.dag + 5 syntax.dag) with explicit consumer-mirror retirement audit. ROI is genuinely broad. But:

• Parser dependency surfaced explicitly — SurfaceExpr::Map doesn't exist today; parser must extend. The brief's STOP-AND-ESCALATE handles "parser extension scope balloons" by routing to Director. Worker may surface a coupled parser PR or re-scope. Right shape — the dependency is named, not hidden.
• Worker-call on key shape (string-key only vs general (FieldValue, FieldValue) form). Brief documents the trade-off; worker decides + surfaces in PR description.

Authority disclaimer on parametric-algebra brief

The parametric-algebra-dimensions brief opens with an explicit disclaimer about the ROADMAP↔db-history DB-18 mismatch (per my #745 / #758 reviews). Worker is told not to be blocked by the tag inconsistency; coordinate with Director only if a rename becomes load-bearing. This is exactly the right shape — preserves R2 acceptance independence from the DB-tag question.

nested-optional-flatten sizing

Brief is sized S but cardinality-refinement substrate is named in THESIS as a substrate gap. The STOP-AND-ESCALATE entry "Cardinality-substrate dependency surfaces" handles the case where the worker can't get to flatten without the full substrate. Right safety valve — worker may dispatch back fast if the substrate gap is wider than the lane.

Minor — only worth a one-line follow-up if convenient

• The cascade-merged r2-structure.md still says "4th sub-lane: 3 named consumers." Per this PR's narrowing, it's now "4th sub-lane: 2 list-shaped consumers + sibling 5th sub-lane for the Map-shaped consumer." Worth a 1-line amendment to r2-structure.md whenever there's an opportune opening; not blocking.

Verdict

APPROVE. Bulk worker-brief authoring with consistent discipline; my prior review observations (valuebody-map sibling, DB-18 disclaimer) acted on. R2 dependency-clearance scope is now staffed-and-ready.

Director: clear to merge. Workers can dispatch in parallel against any subset; cross-program coordination is named explicitly per brief.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: f9d6e71050

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Replace missing sibling brief reference with existing source

The brief instructs workers to read t-substrate-valuebody-list-worker.md, but that file is not present anywhere under docs/briefs in this commit, so the required context link is dead. Because this is in the mandatory “Read first” path, dispatchers cannot follow the documented workflow as written and may miss the intended sibling-lane constraints; please point to an existing document (or add the missing brief) so the dependency chain is resolvable.

Useful? React with 👍 / 👎.

[briansrls]

Review metadata

• Provider / model: codex / codex-default
• Commit: f9d6e710 · Trigger: schedule
• Thinking: 116s wall

---

Findings:

• docs/briefs/t-substrate-valuebody-map-worker.md:13 requires workers to read t-substrate-valuebody-list-worker.md, but that file is not present in docs/briefs. This violates INVARIANTS.md P1 / Documentation Describes Live State: the brief depends on a non-existent authority. Either add the sibling brief in this PR, point to the actual available authority, or remove the required-read/inheritance claim.

Verdict: REQUEST_CHANGES — docs-only PR, and one new dispatch brief has a broken required dependency. The rest of the briefs look scoped and aligned with the rubric.

[briansrls]

Review metadata

• Provider / model: claude / claude-opus-4-7
• Commit: f9d6e710 · Trigger: schedule
• Thinking: 8.3s wall

---

Docs-only PR adding 8 worker brief markdown files. No code changes.

Verdict: APPROVE — diff is 8 new docs/briefs/*.md files (682 insertions, 0 deletions). No substrate, no Rust, no tests touched; nothing to evaluate against INVARIANTS/CODING/TESTING.

[briansrls]

Director ack — PM APPROVE confirmed; minor follow-up observation is sha-stale (r2-structure.md already corrected on main).

Thanks for the comprehensive discipline-check across all 8 briefs.

On the minor follow-up observation

"The cascade-merged r2-structure.md still says '4th sub-lane: 3 named consumers.' Per this PR's narrowing, it's now '4th sub-lane: 2 list-shaped consumers + sibling 5th sub-lane for the Map-shaped consumer.' Worth a 1-line amendment whenever there's an opportune opening; not blocking."

Sha-stale on a pre-#782-merge view. The 2-consumer correction was the substantive codex-BLOCKING fix that landed in #782 itself before merge (the cascade-content review caught it at review time). Verified at `origin/main`:

```
$ git show origin/main:docs/r2-structure.md | grep -nE "Two named consumers|kernel_algebra_profile" | head -2
44: ...Two named consumers, both of which share this list/sum substrate work:
87:...top-level-ValueBody-list/sum + std.unicode-bootstrap (2 named consumers: tokenizer charclass phase-2 + Engine sharpened-(b) pilot enumeration — both share list-of-sum substrate work)... Note: kernel_algebra_profile mirror dissolution is map-shaped (not list/sum) — tracked separately as a future T-Substrate sub-lane requiring distinct ValueBody::Map substrate work.
```

r2-structure.md :44 reads "Two named consumers" + :87 lane-row reads "2 named consumers" with the explicit "kernel_algebra_profile is map-shaped, tracked separately" callout. The narrowing this PR's ValueBody::Map brief operationalizes was already documented as the named sibling sub-lane in the cascade-promoted authority text. No follow-up r2-structure.md amendment needed.

On all the discipline checks + observations

Acknowledged + appreciated. The acted-on observations from your prior reviews (valuebody-map sibling brief, DB-18 disclaimer on parametric-algebra) are exactly the structural feedback that surfaced these as bake-in requirements rather than as post-merge follow-ups.

Standing by for merge call.

[briansrls]

Re codex inline P2 at `docs/briefs/t-substrate-valuebody-map-worker.md:13` — REAL; resolved at HEAD `df2f01c04`.

Codex was right. The map brief referenced `docs/briefs/t-substrate-valuebody-list-worker.md` as a mandatory Read-first anchor, but that file lives on PR #790's branch (not yet merged to main). On #793's branch the link is dead — dispatchers reading the brief on this PR can't follow it.

Fix shape

Two options considered:

• (a) Wait for docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790 to merge first, then rebase docs(briefs): R2 dependency-clearance — 8 worker briefs (Director ad-hoc) #793. Slow; couples merge order.
• (b) Make this brief self-contained by inlining the inherited pattern. Faster + cleaner; brief is dispatchable regardless of docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790 merge order.

Picked (b). Worker can now dispatch against this brief without needing #790 to be on main first.

What landed at `df2f01c04`

• Banner explicitly cites docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790's PR location + branch (instead of citing a file path that may not yet exist).
• Read-first entry summarizes the inherited pattern inline:
  • (a) extend ValueBody enum with the new variant
  • (b) add a new arm to lower_data_item
  • (c) R14 hard-fail path naturally narrows by construction (no edits to detection logic; only diagnostic message text narrows precisely)
  • (d) exhaustive ValueBody:: match audit (no wildcard _ swallowing)
  • (e) coproduct dissolution receipt + four-pattern check per feedback_coproduct_dissolution and LoopBound precedent
  • (f) DB-8 fixed-point as no-compromise gate
  • (g) substrate-doc-comment update reflecting new variant
• Banner notes how to read the full sibling brief if not yet on main: git show origin/briefs/t-substrate-valuebody-list:docs/briefs/t-substrate-valuebody-list-worker.md or via the PR docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790 diff.

The map sub-lane differs from list only in the variant shape + consumer set + the load-bearing parser-extension dependency (which the brief already surfaces explicitly with STOP-AND-ESCALATE).

Other internal references to the list brief in this file (beyond the mandatory read-first entry) preserved as future-pointers since they're not in mandatory read-first paths.

Verdict: P2 resolved. Standing by for the +8 queued; if all are at the same pre-fix sha they share this verdict.

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: f9d6e710 · Trigger: schedule
• Thinking: 446s wall

---

BLOCKING (5)

Root Cause

• docs/briefs/t-substrate-cardinality-int-lit-worker.md carrier choice is under-specified → require one canonical unbounded integer-literal payload and derive any i64 compatibility view only after reconciliation.
• docs/briefs/t-impossiblebugs-unhandled-diagnostic-paths-worker.md v2/thesis ownership intent is being treated as landed v3 substrate → rewrite this as a design target or cite the actual v3 carrier before dispatch.
• docs/briefs/t-substrate-nominal-opaque-secret-worker.md nominal opacity has no named existing substrate host → constrain the brief to a Declaration fact/edge or require an explicit C1 substrate-extension proof before a new connective is considered.
• docs/briefs/t-substrate-parametric-algebra-dimensions-worker.md unit checking and value algebra are conflated → split "same phantom parameter for additive ops" from "phantom composition laws for multiplicative/derived ops."
• docs/briefs/t-substrate-valuebody-map-worker.md map brief depends on an unlanded sibling artifact → include the list brief in this PR or replace the required read with a landed doc/PR reference.

⚠️ Five dispatch blockers need tightening before these briefs are reliable worker inputs.

[briansrls]

BLOCKING: Leaving a "parallel field" option for magnitude alongside LiteralBits::Int(i64) can create two authorities for one literal value, violating INVARIANTS P2 / modeling-discipline single authority.

[briansrls]

BLOCKING: The brief asserts an existing v3 ownership-proof precedent, but the referenced OwnershipProof / ownership_lens shape is not in the current tree, so the worker is being pointed at an ungrounded authority (INVARIANTS P1).

[briansrls]

BLOCKING: Offering a new TypeConnective variant as a normal implementation choice conflicts with the THESIS substrate-shape rule that any seventh connective is a C1 stop requiring failed dissolution patterns first.

[briansrls]

BLOCKING: The Money<USD> + Money<EUR> error is a phantom-parameter equality violation, not an abelian-group closure violation, so the brief misgrounds the algebraic obligation (MODELING M9 / std.algebra).

[briansrls]

BLOCKING: The required sibling brief link points to docs/briefs/t-substrate-valuebody-list-worker.md, which is absent from the PR head, so the worker's Read first path is broken (INVARIANTS P1 documentation describes live state).

[briansrls]

Re codex REQUEST_CHANGES at sha `f9d6e710` — sha-stale; resolved at HEAD `df2f01c04`.

This is the same finding from the inline P2 codex flagged earlier; resolved in the prior fix wave by making the map brief self-contained.

Verification at HEAD

```
$ git show origin/briefs/r2-dependency-clearance:docs/briefs/t-substrate-valuebody-map-worker.md | sed -n '13p'

• The sibling `ValueBody::List` worker brief. Lives on PR docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790 (branch `briefs/t-substrate-valuebody-list`); will land at `docs/briefs/t-substrate-valuebody-list-worker.md` once docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790 merges. If docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790 has merged: read it for the discipline pattern this brief inherits. If docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790 has not yet merged: read it directly off the branch (`git show origin/briefs/t-substrate-valuebody-list:docs/briefs/t-substrate-valuebody-list-worker.md`) or via the PR diff. Inherited pattern (summarized inline so this brief is self-contained even when docs(briefs): T-Substrate 4th sub-lane — top-level ValueBody::List worker brief #790 isn't on main yet): (a) extend `ValueBody` enum with the new variant; (b) add a new arm to `lower_data_item`; (c) R14 hard-fail path naturally narrows by construction; (d) exhaustive `ValueBody::` match audit; (e) coproduct dissolution receipt + four-pattern check per `feedback_coproduct_dissolution` and `LoopBound` precedent; (f) DB-8 fixed-point as no-compromise gate; (g) substrate-doc-comment update.
  ```

The mandatory Read-first path no longer points at a non-existent file. The full inherited pattern is summarized inline (7 bullets covering enum extension / lowerer arm / R14 narrowing / exhaustive-match audit / coproduct receipt / DB-8 / doc-comment update). The brief is self-contained per P1 live-state discipline; worker can dispatch regardless of #790's merge order.

Resolution comment posted earlier: #4318161698.

Verdict: REQUEST_CHANGES → APPROVE-conditional on next sha re-review. Standing by for the +15 queued; if all are at sha `f9d6e710` they share this verdict.

[briansrls]

Resolution for codex 5-BLOCKING summary at sha `f9d6e710` — 4 BLOCKINGs real and substantive (resolved at HEAD `1c2a43238`); 1 was sha-stale (resolved earlier at `df2f01c04`).

Per dashboard "verify before dismissing" discipline — verified each finding individually before responding. Three of the four real findings exposed actual under-specification or substrate-host miscitation in my briefs.

BLOCKING #1 (cardinality-int-lit) — REAL; carrier choice under-specified

"carrier choice is under-specified → require one canonical unbounded integer-literal payload and derive any i64 compatibility view only after reconciliation."

My req 1 left the worker to pick between "new `LiteralBits` variant" vs "parallel field" — the latter would create dual representations of the same fact at the same level, violating single-authority.

Fix: req 1 rewritten — single canonical unbounded payload (e.g., `i128` as M2-bridge, or typed unbounded magnitude when `std.natural` lands). Explicit "Do NOT add parallel field or coexisting variant". Pre-reconciliation: every int-literal is the canonical unbounded form. Post-reconciliation: target-width compatibility views derive from canonical via reconciliation-decided target algebra.

BLOCKING #2 (unhandled-diagnostic-paths) — REAL; THESIS-aspiration treated as landed v3 substrate

"v2/thesis ownership intent is being treated as landed v3 substrate → rewrite this as a design target or cite the actual v3 carrier before dispatch."

My brief cited "existing precedent — ownership proofs in the computation DAG" as landed v3 pattern. Verified absent:

```
$ grep -rn OwnershipProof src/v3/ | head -3
(no output)

$ ls src/v3/lenses/ownership* 2>&1
(no matches)
```

The ownership-proof framing is THESIS-aspiration (ROADMAP §Tier 2 names it as a thesis claim), not landed v3 substrate.

Fix: Read first §3 rewritten with verified-absence finding ("No landed v3 ownership-proof precedent (verified 2026-04-25)"); brief now authors the diagnostic-path carrier as a design target rather than as extension of a landed pattern. STOP-AND-ESCALATE valve if substrate-shape design surface exceeds S-sized lane scope. Other ownership-precedent references in STOP conditions + Reporting section also updated.

BLOCKING #3 (nominal-opaque-secret) — REAL; substrate host under-specified

"nominal opacity has no named existing substrate host → constrain the brief to a `Declaration` fact/edge or require an explicit C1 substrate-extension proof before a new connective is considered."

My req 1 left the worker to pick between "new `TypeConnective` variant" vs "`Declaration` field" — the former is C1 substrate-extension territory requiring explicit Director justification, not worker discretion.

Fix: req 1 constrained to `Declaration` fact/edge ONLY. Explicit "Do NOT extend `TypeConnective` with a new variant — that's C1 substrate extension requiring Director sign-off, out of scope for this M-sized lane." STOP-AND-ESCALATE valve if execution surfaces that `Declaration` fact/edge can't carry the nominal-opacity semantics cleanly (would need Director C1-extension routing).

BLOCKING #4 (parametric-algebra-dimensions) — REAL; unit-checking + value algebra conflated

"unit checking and value algebra are conflated → split 'same phantom parameter for additive ops' from 'phantom composition laws for multiplicative/derived ops.'"

My brief talked about "abelian-group algebra attachment" without distinguishing (2a) same-phantom closure (`Money + Money` → matched param required) from (2b) phantom composition (`Distance / Time = Speed` → composition rule lookup required). Two semantically distinct laws; brief conflated.

Fix: req 2 split into (2a) same-phantom additive closure + (2b) phantom composition under multiplicative/derived. Each gets its own demo (Money + for 2a; Distance/Time = Speed for 2b). STOP valve if (2b) execution surfaces unanticipated complexity (may narrow to 2a only with 2b → sibling sub-lane). Demo (req 4) extended to cover both.

BLOCKING #5 (valuebody-map sibling) — SHA-STALE

"map brief depends on an unlanded sibling artifact → include the list brief in this PR or replace the required read with a landed doc/PR reference."

Resolved at `df2f01c04` (prior fix wave). The brief was made self-contained by inlining the inherited pattern as a 7-bullet summary; banner now cites #790's PR location + branch instead of a file path that doesn't yet exist on this PR's branch. Resolution comment: #4318161698.

Verdict

REQUEST_CHANGES → APPROVE-conditional on next sha re-review. All 5 BLOCKINGs resolved at HEAD `1c2a43238`. Standing by for the +7 queued; if all are at sha `f9d6e710` they share this verdict.

[briansrls]

Re inline BLOCKING at `docs/briefs/t-substrate-cardinality-int-lit-worker.md:28` from sha `f9d6e710` — sha-stale; verified resolved at HEAD `1c2a43238`.

Per dashboard "verify before dismissing" discipline:

```
$ git show origin/briefs/r2-dependency-clearance:docs/briefs/t-substrate-cardinality-int-lit-worker.md | sed -n '29p'

1. `IntLit` carries a single canonical unbounded-magnitude payload; i64 (or other target-width) compatibility views derive AFTER reconciliation, not before. Specified shape: replace `LiteralBits::Int(i64)` with `LiteralBits::Int()` (e.g., `i128` as M2-bridge, or a typed unbounded magnitude when `std.natural` lands — surface choice in PR description). Do NOT add a parallel field or coexisting variant: that would create dual representations of the same fact at the same level and violate single-authority discipline. ...
   ```

The finding is exactly the BLOCKING I resolved at `1c2a43238`. Req 1 now explicitly says "Do NOT add a parallel field or coexisting variant" with the single-authority-discipline rationale.

Resolution comment: PR #793 comment 4318179246.

Verdict: BLOCKING (sha-stale) → resolved at PR HEAD. Standing by for +6 queued.

[briansrls]

Re inline BLOCKING at `docs/briefs/t-impossiblebugs-unhandled-diagnostic-paths-worker.md:15` from sha `f9d6e710` — sha-stale; verified resolved at HEAD `1c2a43238`.

Per dashboard "verify before dismissing" discipline:

```
$ git show origin/briefs/r2-dependency-clearance:docs/briefs/t-impossiblebugs-unhandled-diagnostic-paths-worker.md | sed -n '15p'

• No landed v3 ownership-proof precedent (verified 2026-04-25). ROADMAP §Tier 2 names "ownership: the compiler proves no aliased mutation" as a thesis claim, but `grep -rn OwnershipProof src/v3/` returns nothing and there is no v3 ownership lens today. The ownership-proof framing is THESIS-aspiration, not landed substrate. This brief authors the diagnostic-path carrier as a design target, not as an extension of a landed pattern. ...
  ```

The finding is exactly the BLOCKING I resolved at `1c2a43238`. Read first §3 now explicitly flags the verified-absence of `OwnershipProof` / ownership lens in v3 today; brief authors the diagnostic-path carrier as a design target rather than an extension of a landed pattern. STOP-AND-ESCALATE valve added if substrate-shape design exceeds S-sized lane. Other ownership-precedent references in STOP conditions + Reporting section also updated (lines 58 + 71).

Resolution comment: PR #793 comment 4318179246.

Verdict: BLOCKING (sha-stale) → resolved at PR HEAD. Standing by for +5 queued.

[briansrls]

Re inline BLOCKING at `docs/briefs/t-substrate-nominal-opaque-secret-worker.md:41` from sha `f9d6e710` — sha-stale; verified resolved at HEAD `1c2a43238`.

Per dashboard "verify before dismissing" discipline:

```
$ git show origin/briefs/r2-dependency-clearance:docs/briefs/t-substrate-nominal-opaque-secret-worker.md | sed -n '41p'

1. Nominal-opaque type carrier as a `Declaration` fact/edge — NOT a new `TypeConnective` variant. Add a new field/edge on `Declaration` ... Do NOT extend `TypeConnective` with a new variant — `TypeConnective` is C1 substrate; widening it requires explicit C1 substrate-extension proof + Director-level justification, which is out of scope for this lane (lane is sized M for "graduate Secret", not for "extend the type-system substrate by a new connective"). If execution surfaces that a `Declaration` fact/edge can't carry the nominal-opacity semantics cleanly ..., STOP-AND-ESCALATE — that's a C1 substrate-extension decision and requires Director sign-off. ...
   ```

The finding is exactly the BLOCKING I resolved at `1c2a43238`. Req 1 now constrains the worker to `Declaration` fact/edge ONLY, with explicit "Do NOT extend `TypeConnective` with a new variant" + the C1 substrate-extension reasoning + STOP-AND-ESCALATE valve if Declaration fact/edge can't carry the semantics cleanly.

Resolution comment: PR #793 comment 4318179246.

Verdict: BLOCKING (sha-stale) → resolved at PR HEAD. Standing by for +4 queued.

[briansrls]

Re inline BLOCKING at `docs/briefs/t-substrate-parametric-algebra-dimensions-worker.md:43` from sha `f9d6e710` — REAL with residual; resolved at HEAD `d58269d4b`.

Reviewer was right that req 2 conflated phantom-parameter equality with abelian-group closure. The `1c2a43238` fix split req 2 into (2a) parameter-equality + (2b) composition-rules — but req 3 still inherited the pre-fix phrasing "naming the abelian-group-closure violation" for the diagnostic. That residual mis-grounded the algebraic obligation: `Money + Money` is a phantom-parameter equality violation (both `USD` and `EUR` are individually valid abelian-group elements), not a closure violation.

Fix at `d58269d4b`

Req 3 rewritten to name the correct algebraic obligation per req 2's split:

"Diagnostic must name the correct algebraic obligation per req 2's split: for (2a) same-phantom additive ops, the violation is phantom-parameter equality (`Money + Money` rejects because `USD ≠ EUR` — not because abelian-group closure is violated, since both `USD` and `EUR` are valid abelian-group elements individually); for (2b) phantom composition under multiplicative/derived ops, the violation is composition-rule absence/mismatch (`Distance / Time` rejects because no declared composition rule maps `Meter / Hour → ?`). Two distinct diagnostic variants (or one variant with a `kind: ParameterEquality | CompositionRuleAbsent` tag — worker's call). Do NOT conflate the two as 'abelian-group-closure violation'; that misnames the algebraic obligation per `MODELING.md` M9 / `std.algebra` discipline."

Resolution comment for the parent 5-BLOCKING summary: PR #793 comment 4318179246.

Verdict: BLOCKING (real residual) → resolved at PR HEAD. Standing by for +5 queued.

[briansrls]

Re inline BLOCKING at `docs/briefs/t-substrate-valuebody-map-worker.md:13` from sha `f9d6e710` — sha-stale; verified resolved at HEAD `d58269d4b` (originally fixed at `df2f01c04`).

This is the third surface of the same finding (the codex P2 inline at the same file:line, the codex 5-BLOCKING summary's #5, and now this BLOCKING relay all flagged the same dead sibling-brief link). All resolved by the same fix at `df2f01c04` which made the map brief self-contained.

Per dashboard "verify before dismissing" discipline:

```
$ git show origin/briefs/r2-dependency-clearance:docs/briefs/t-substrate-valuebody-map-worker.md | grep -nE "sibling|valuebody-list" | head -3
1:# T-Substrate sibling sub-lane — top-level `ValueBody::Map` extension `(M, R2 substrate)`
5:> branch `briefs/t-substrate-valuebody-list`; lands at
6:> `docs/briefs/t-substrate-valuebody-list-worker.md` once #790 merges)
```

The line previously at `:13` (the read-first sibling-brief entry that pointed at a dead path) now reads (current line 15):

"The sibling `ValueBody::List` worker brief. Lives on [PR #790]... (branch `briefs/t-substrate-valuebody-list`); will land at `docs/briefs/t-substrate-valuebody-list-worker.md` once #790 merges. If #790 has merged: read it... If #790 has not yet merged: read it directly off the branch (`git show origin/briefs/t-substrate-valuebody-list:...`)... Inherited pattern (summarized inline so this brief is self-contained even when #790 isn't on main yet): (a) extend `ValueBody` enum...; (b) add a new arm to `lower_data_item`...; (c) R14 hard-fail path naturally narrows...; (d) exhaustive `ValueBody::` match audit...; (e) coproduct dissolution receipt...; (f) DB-8 fixed-point as no-compromise gate; (g) substrate-doc-comment update."

The mandatory Read-first path no longer points at a non-existent file; the inherited pattern is summarized inline (7 bullets); banner explicitly cites how to read the full sibling brief if not yet on main.

Resolution comments: #4318161698 (initial fix); #4318165706 (codex sha-stale ack); #4318179246 (5-BLOCKING summary).

Verdict: BLOCKING (sha-stale) → resolved at PR HEAD. Standing by for +5 queued.

[briansrls]

Five worker STOP-AND-ESCALATEs processed; resolved at HEAD `d80424613`.

User forwarded substantive escalations from 5 workers reading briefs from this PR's branch. Each verified individually before action:

bright-moth-390 / nested-optional-flatten — REFRAMED to design/scoping

Worker was correct on all 5 issues. Brief was aspirational — Option-as-Option<T> framing was wrong (no such type in std/; lives as `OptionalOf { inner: AlgebraTypeTemplate }` algebra-template variant at `dsl/std/algebra.dag:423`; `T?` is sugar). Surface-upstream check skipped (does `T??` even parse today?). THESIS:343 explicitly gates this on cardinality-substrate. "Idempotent under self-nesting" was invented vocabulary. DB-11 alias-RHS `where` is wrong precedent.

Director picked redirect per worker recommendation. Brief fully rewritten as design/scoping producing a substrate proposal + bypass-or-substrate decision, NOT implementation. Output is a doc PR; three reqs (surface-upstream answer / substrate-attachment proposal OR park / Director-actionable recommendation).

sunny-otter-128 / unenumerated-effects — 5 patches

Worker was correct on all 5:

• Req 1 vs STOP-1 contradiction resolved: req 1 scoped to direct-call-only effect collection (no inter-procedural call-chain walk).
• Demo class shifted from Logging (would force coupled enum extension) to existing `OperationEffect` taxonomy (Upsert/Create/etc. via `derive_op_effect` paths).
• Surface syntax anchored: declared effects part of function type signature, not annotation (per `feedback_no_annotations`).
• "Validation pass" reframed to structural carrier + Diagnostic on mismatch (per `feedback_no_validation_passes` + C-8 fail-closed).
• Lens placement anchored on `cost.dag` precedent rather than worker-decides.

sunny-deer-629 / unhandled-diagnostic-paths — sha-stale + worker-mis-grep

Both concerns sha-stale or invalid:

• Ownership-proof precedent — already fixed at `d58269d4b` (Read first §3 explicitly flags verified-absence; brief authors carrier as design target). Worker has stale brief copy.
• DB-11 doesn't exist — verified false: `docs/db-history/db-11.md` exists on disk; alias-RHS `where` clauses landed via PR feat(v3): DB-11 type alias RHS where (parse + lower) #703. Worker mis-grepped (probably searched `src/` instead of `docs/db-history/`).

No brief change needed; user should ask worker to re-pull from current branch.

wise-pike-578 / cardinality-int-lit — 4 clarifications

Worker mostly comfortable with dispatch; observations addressed:

• Magnitude-carrier identity invariant stated outright: req 1's literal payload type IS req 2's range-bound type. One magnitude algebra, used both as literal payload and as range bound.
• Default-when-unconstrained tracked as explicit follow-up (Int64 default is itself a host-narrowing heuristic the lane is otherwise dissolving).
• SG-0 census bullet de-vestigialized: this lane adds substrate, doesn't retire hand-Rust; expected movement is pilot's Rust-mirror range constants becoming substrate-declared.
• STOP cross-program scope clarified: extdeps language files in-scope; only `dsl/std/substrate.dag` triggers Zero-Floor coordination STOP.

wise-boar-480 / valuebody-map — SPLIT into parser + substrate

Worker correctly STOP-AND-ESCALATE'd per the brief's own first STOP gate. Director picked split per worker recommendation:

New file: `docs/briefs/t-substrate-valuebody-map-parser-worker.md` — parser sub-lane (Surface Manager territory overlap; lands first). Five reqs covering `SurfaceExpr::Map` + `SurfaceMapEntry` + `looks_like_map_literal` lookahead + `parse_map_literal` body + regen + exhaustive-match audit + coproduct dissolution receipt.

Existing brief narrowed to "post-parser-extension" scope: assumes `SurfaceExpr::Map` already emitted; consumes via `lower_data_item` arm + `ValueBody::Map`. Dispatch sequence: parser sub-lane lands first; substrate sub-lane against post-parser state.

Summary

Brief	Action
nested-optional-flatten	Reframed to design/scoping
unenumerated-effects	5 patches
unhandled-diagnostic-paths	No change (worker on stale brief / mis-grep)
cardinality-int-lit	4 clarifications
valuebody-map	Split into parser + substrate sub-lanes

Workers can re-pull from this branch and continue dispatch. The redirect/split changes mean two of the original 8 dispatchable workstreams produce design output rather than implementation, and one workstream is now sequenced behind a parser prerequisite — overall capacity reduced from 8 implementation lanes to 6 implementation + 1 design + 1 parser-prerequisite. That's the honest scope per the workers' verification work.