chore(b-0139): decompose MEMORY.md backfill into B-0527#3315
Conversation
![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: 68732ffd5a
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".
There was a problem hiding this comment.
Pull request overview
Decomposes the B-0139 MEMORY.md backfill work into a dedicated backlog row, B-0527, and updates the parent row status to reference that slice.
Changes:
- Adds a new P1 backlog row for MEMORY.md backfill work.
- Updates B-0139 status text to point the MEMORY.md backfill slice at B-0527.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md |
Adds the new atomic backlog row for B-0139 MEMORY.md backfill. |
docs/backlog/P1/B-0139-pre-substrate-kenji-era-otto-work-inventory-aaron-2026-05-01.md |
Updates parent backlog status text to mention B-0527. |
Comments suppressed due to low confidence (6)
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md:7
- This new backlog row is missing the required
last_updatedfrontmatter field. The backlog schema markslast_updatedas required and says it must be updated on every content edit, so the row should include the creation date here.
created: 2026-05-14
depends_on: [B-0522, B-0523, B-0526]
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md:4
not-startedis not one of the backlog schema status values (open,closed,superseded-by-B-NNNN,deferred,decomposed). New backlog rows that are not yet in progress should use the schema's open-state value so tooling and audits don't have to handle a new unregistered status.
status: not-started
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md:7
- This dependency list references
B-0526, but no backlog row withid: B-0526exists indocs/backlog. That leaves this new row with a dangling prerequisite and makes the dependency order impossible to satisfy until the referenced row is added or corrected.
depends_on: [B-0522, B-0523, B-0526]
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md:16
- This row is described as decomposed from
B-0139, but it does not include theparent: B-0139frontmatter used by sibling decomposed rows likeB-0522andB-0523. Without that parent link, parent/child searches and decomposition audits will miss B-0527.
**Filed by:** Lior (decomposed from B-0139 blob)
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md:22
- The prerequisite descriptions don't match the referenced rows:
B-0522is the git branches/worktrees inventory, not an F#src/Coreinventory, andB-0526is not present in the backlog. As written, implementers won't know which row actually produces the F# artifact inventory this backfill depends on.
This composes with task #291 (MEMORY.md index audit + backfill). It relies on the artifacts identified by B-0522 (F# src/Core inventory), B-0523 (docs/research cross-reference audit), and B-0526 (branch/worktree content inventory).
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md:26
- The acceptance criterion includes
B-0526, but that backlog ID does not exist indocs/backlog. This makes the completion condition depend on an undefined source of artifacts, so the criterion should be corrected or the missing prerequisite row added.
1. Every identified pre-substrate Kenji-era artifact (from B-0522, B-0523, B-0526) without a substrate reference has an appropriate pointer added to `MEMORY.md` (or the relevant persona memory index).
…OG.md - Change status from not-started to open (matches documented values) - Add parent: B-0139 field for decomposition tracking - Add last_updated: 2026-05-14 - Remove B-0526 from depends_on (parallel sibling, not a prerequisite) - Add B-0527 to B-0139 children list - Add B-0527 entry to docs/BACKLOG.md P1 section Co-Authored-By: Claude <noreply@anthropic.com>
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: a68de6f1a9
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".
The audit-formal-artifacts.ts script covers Lean/TLA+/Z3/Alloy but not F# src/Core. Restoring this slice explicitly per acceptance criterion 2 to prevent B-0139 appearing completed while src/Core is still untracked. Co-Authored-By: Claude <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.
Comments suppressed due to low confidence (3)
docs/backlog/P1/B-0139-pre-substrate-kenji-era-otto-work-inventory-aaron-2026-05-01.md:76
- This status introduces a dangling backlog reference:
B-0526does not exist anywhere underdocs/backlog, while the existing branch/worktree child isB-0522. Please either add the missing row and include it in the decomposition metadata, or update this reference to the existing child so the B-0139 decomposition graph is navigable.
**In progress.** First slice landed: `tools/hygiene/audit-formal-artifacts.ts` — a TS script (Rule 0 compliant) that catalogs all formal verification artifacts (Lean4, TLA+, Z3, Alloy, formal tests) and cross-references each against docs/ for substrate-status. Finds 30 artifacts (4295 lines); 24 referenced in substrate, 6 unreferenced TLA+ specs (`AsyncStreamEnumerator.tla`, `BftConsensus.tla`, `ChaosEnvDeterminism.tla`, `ConsistentHashRebalance.tla`, `FeatureFlagsResolution.tla`, `InfoTheoreticSharder.tla`). Remaining slices: docs/research/ cross-reference audit (decomposed to B-0523), branch/worktree content inventory (decomposed to B-0526), F# src/Core built-artifact catalog (acceptance criterion 2 — not yet covered by audit-formal-artifacts.ts), MEMORY.md backfill (decomposed to B-0527).
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md:24
- The dependency description is inconsistent with the actual backlog rows:
B-0522is the git branches/worktrees inventory, not the F#src/Coreinventory, andB-0526is not present indocs/backlogordocs/BACKLOG.md. This leaves the new MEMORY.md backfill row pointing at the wrong/missing prerequisites.
This composes with task #291 (MEMORY.md index audit + backfill). It relies on the artifacts identified by B-0522 (F# src/Core inventory), B-0523 (docs/research cross-reference audit), and B-0526 (branch/worktree content inventory).
docs/backlog/P1/B-0527-memory-md-backfill-pre-substrate-kenji-era.md:28
- This acceptance criterion is not aligned with the parent row's scope: it references missing
B-0526, and it does not include the F#src/Corebuilt-artifact catalog that B-0139 still lists as an uncovered prerequisite. As written, B-0527 could be closed without backfilling all artifacts B-0139 requires.
1. Every identified pre-substrate Kenji-era artifact (from B-0522, B-0523, B-0526) without a substrate reference has an appropriate pointer added to `MEMORY.md` (or the relevant persona memory index).
…dren B-0526 (branch/worktree content inventory) was merged to main via PR #3309 before this PR was cut — making the body-text references to B-0526 in B-0527 correct but inconsistent with the frontmatter that had removed it. - B-0527 depends_on: [B-0522, B-0523] → [B-0522, B-0523, B-0526] - B-0139 children: [B-0522, B-0523, B-0527] → [B-0522, B-0523, B-0526, B-0527] Resolves copilot/codex unresolved threads on PR #3315. Co-Authored-By: Claude <noreply@anthropic.com>
|
Closing as redundant — substrate already on This PR was decomposing B-0139 (MEMORY.md backfill) into a new atomic backlog row B-0527. A different PR shipped the same intent under a more descriptive filename: Per Note: B-0527 ID was also claimed by PR #3323 ("protocol doc scaffold") — a separate ID-allocation collision documented in Substrate-honest, no force-merge needed. — otto-cli @ 2026-05-23 |
Pull request was closed
Decomposes the MEMORY.md backfill slice of the B-0139 blob into a dedicated atomic row B-0527.