chore(repo): housekeeping for PR #124 — NS-16 fold + Plan-007 flip#125
Merged
Conversation
Follow-on housekeeping for the Tier-4 plan-readiness audit (PR #124, squash `cb3d1b2`), per the multi-PR housekeeping discipline that defers the §6 status fold out of the audit-swap PR. §6 fold (cross-plan-dependencies.md) — mirrors the Tier-3 precedent (030c4ec): - NS-16 (Tier 4) flips `ready` → `completed`; the Mermaid node splits into a `:::completed` NS-16 node + a new collapsed `:::ready` NS-17 node, joined by the `NS-16 --> NS-17` audit-chain edge. - NS-17 (Tier 5) promoted `blocked` → `ready`; NS-18..NS-21 stay `blocked` behind it (strict serialization). Status / Priority / Upstream / Summary detail lines + the tier-PR checklist (tier-3 de-staled to PR #118, tier-4 checked → PR #124) + the §6 narrative amendment updated. Plan-007 status `approved` → `review` — applies the mandatory flip PR #124 ratified in REVIEW.md but never wrote to the file. The Tier-4 audit reopened the Phases 4–7 design (3 new Phases R1/R2/R3 + 8 new CPs CP-007-7..14 + 12 new invariants I-007-8..19), which triggers the runbook §Status Flip Rule. The `review` state is the load-bearing downstream-gate signal, and the tracking state for the contract finalizations still pending (error-contracts `daemon.lifecycle_conflict` mapping; §3 Plan-007 deps annotation) that land in Plan-007's review→approved cycle / R1 task T-007r-1-9. Plan-005 + Plan-006 stay `approved` (additive amendments only). The Phases 1–3 partial (PRs #16/#17/#19) is preserved in-plan as audit evidence and is unaffected. Validation: prettier --check + lychee (--offline, 0 errors) + docs-corpus pre-commit-runner (exit 0; mermaid-set-coherence green). Refs: Plan-005, Plan-006, Plan-007 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: d35051270d
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".
Both findings target ripples of this PR's Plan-007 `approved` → `review` flip and
the NS-16 §6 fold; neither was caught by the static mermaid-set-coherence validator
(it checks the narrative's own set-claims against node classes — both authored here).
P2 (cross-plan-dependencies.md §NS-15..21 Summary) — the flip to `review` made the
pre-existing "Plan-017 — the only `review`-status plan" clause contradictory. Reworded
to name both review-status plans and preserve the distinction: Plan-017 is pending-audit
`review` (its Tier-8 audit resolves it); Plan-007 is post-audit-reopen `review` (its own
R1/R2/R3 cycle resolves it, not a pending tier audit). Closes the governance-pointer
gap that could let the reopened Plan-007 promotion cycle be missed.
P2 (cross-plan-dependencies.md §6 Mermaid) — the collapsed `NS-17..NS-21` node was
classed `:::ready`, presenting blocked Tier 6-9 audits as dispatchable. Narrowed NS17
to Tier 5 only (mirroring the NS16 node format) + added an explicit
`NS18[NS-18..NS-21]:::blocked` node + the `NS17 --> NS18` chain edge. The graph now
matches the narrative blocked set `{NS-09, NS-10, NS-18..NS-21}` and the §NS-15..21
Status line, removing the graph-vs-text mismatch.
Validation: prettier --check + lychee (--offline, 0 errors) + docs-corpus pre-commit-runner
(exit 0; mermaid-set-coherence green on the split).
Refs: Plan-007, Plan-017
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Owner
Author
|
@codex review |
|
Codex Review: Didn't find any major issues. Nice work! ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
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". |
The Plan-007 `approved` -> `review` flip (this PR, d350512) made two README status assertions stale on lines the diff never touched — invisible to diff-scoped Codex review and the link/anchor/mermaid validators. Surfaced by a post-flip ripple sweep (prompted by Codex catching the analogous §6 line 542). Measured the actual status distribution from docs/plans + docs/specs `Status` fields rather than inheriting README's own counts (which proved unreliable): - Plans: 24 `approved` + 1 `completed` (Plan-001) + 2 `review` (Plan-007, Plan-017) = 27. README's prior "26 approved" had silently miscounted Plan-001 as approved (it is `completed`). Corrected line 216 prose + line 259 bullet. - Specs: 27 `approved` (all). README's "26 approved + Spec-027 in draft" was stale since NS-13b's 2026-05-25 Spec-027 `draft` -> `approved` promotion — pre-existing, independent of this flip, fixed as an adjacent drive-by found in the same sweep (line 260). Deliberately scoped to the two counts with direct evidence (plans = this flip's ripple; specs = verified adjacent). The block's other counts (domain models, architecture docs, runbooks, ADRs) have no staleness signal and were left untouched. Refs: Plan-001, Plan-007, Plan-017, Spec-027 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Owner
Author
|
@codex review |
|
Codex Review: Didn't find any major issues. Chef's kiss. ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
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". |
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.
Summary
Follow-on housekeeping for the Tier-4 plan-readiness audit (#124, squash
cb3d1b2), per the multi-PR housekeeping discipline that lands the §6 status fold + the ratified status flip out of the audit-swap PR.1. §6 fold (
cross-plan-dependencies.md) — completes NS-16, mirroring the Tier-3 precedent (030c4ec):ready→completed; the Mermaid node splits into a:::completedNS-16 node + a new:::readyNS-17 node, joined by theNS-16 --> NS-17audit-chain edge.blocked→ready; NS-18..NS-21 stayblockedbehind it (strict serialization).{NS-16}→{NS-17}).2. Plan-007 status
approved→review— applies the mandatory flip PR #124 ratified in the audit REVIEW.md but never wrote to the file. The Tier-4 audit reopened the Phases 4–7 design (3 new Phases R1/R2/R3 + 8 new CPs CP-007-7..14 + 12 new invariants I-007-8..19), triggering the runbook §Status Flip Rule. Plan-005 + Plan-006 stayapproved; the shipped Phases 1–3 partial (#16/#17/#19) is preserved in-plan as audit evidence.3. README status-census reconciliation — ripple of change 2; found by the post-flip sweep (see § README census reconciliation).
Codex P2 fixes (commit
e3695cb)Codex's review of the §6 fold (
d350512) surfaced two P2 ripples of the flip — neither catchable by the static mermaid-set-coherence validator (it checks the narrative's own set-claims against node classes, both authored here):review-status plan" clause false. Reworded to name both review-status plans + their distinct promotion paths (Plan-017 pending-audit; Plan-007 post-audit-reopen).NS-17..NS-21:::readynode presented blocked Tier 6-9 audits as dispatchable. Split intoNS-17(Tier 5,:::ready) +NS-18..NS-21(:::blocked) + theNS-17 --> NS-18chain edge, matching the narrative blocked set.README census reconciliation (commit
0b1266b)Codex catching the §6 line-542 set-claim falsified an earlier "no other corpus assertion" ripple conclusion — so the ripple sweep was re-run comprehensively (diff-scoped review + the link/anchor/mermaid validators are all structurally blind to status-claim semantics on untouched lines). It found three stale README status lines, with the counts measured from
docs/plans+docs/specsStatusfields rather than inherited from README's own numbers (which proved unreliable):reviewwhile the other V1 plans are approved"completed, Plan-007 + Plan-017review, restapprovedapprovedand Plan-017 is inreview"approved, Plan-001 iscompleted, and Plan-007 + Plan-017 are inreview"draft"draft→approved(pre-existing, drive-by)Measured distribution: plans = 24
approved+ 1completed(Plan-001) + 2review(Plan-007, Plan-017) = 27; specs = 27approved(all). README's prior "26 approved" plans had silently miscounted Plan-001 (completed, notapproved) — independent of this flip but corrected here since the same lines were being edited. Deliberately scoped to the two counts with direct evidence; the block's other counts (domain models, arch docs, runbooks, ADRs) have no staleness signal and were left untouched.Post-merge #124 reconciliation
REVIEW.md's ratified change-manifest was cross-referenced row-by-row against the merged corpus (verified by diff/grep, not recall — REVIEW.md is pre-execution intent; #124's 6 Codex rounds are the newer authority and superseded several manifest items on purpose). The status flip is the one confirmed, load-bearing miss; every other manifest item resolved to already-landed, reconciled-away, or no-op:
reviewDaemonStop/Restart+data.type: daemon.lifecycle_conflictenvelope +providerFailureDetailDaemonHandshake*local-ipc-gateway.ts(shipped), not a new R1 schema:::ready" + test-plan L207030c4ec) likewise never flipped namespace lanes.daemon-supervision/→daemon-status-projector/daemon-supervisionpath exists in §6 to renameTwo genuine residuals — surfaced with a recommended disposition (not passively deferred)
Both verified absent from the merged corpus. Neither belongs in a §6 status-fold PR (each is a contract-doc / dependency-graph edit with its own review surface), and both are owned by Plan-007's now-reopened R1/R3 design:
error-contracts.md§Plan-007 Tier 4 Domain Identifiers —daemon.lifecycle_conflict→ JSON-RPC-32603(+ supervision-disconnect reason taxonomy)-32603is the established InternalError classcross-plan-dependencies.md§3 Plan-007 §Dependencies row — annotate CP-007-7..14 (esp. R3 ↔ Plan-023 Tier-8 renderer chrome)review→approved. The annotation encodes a Tier-4 → Tier-8 cross-tier dependency resolved by the R3 design re-validation now underway.The decision on opening the dedicated PR (and whether to bundle both residuals or split them) is the maintainer's.
Validation
prettier --check— clean (README.md + cross-plan-dependencies.md; no code-span/italic mangling)lychee --offline— 0 errors (README 30/30 OK; corpus 240/246 OK, 6 excluded)docs-corpus pre-commit-runner— exit 0 (path-canonical-ripple + mermaid-set-coherence + cite-target-existence)Refs: Plan-001, Plan-005, Plan-006, Plan-007, Plan-017, Spec-027
🤖 Generated with Claude Code