decompose(B-0118): smallest atomic children (TS-first, re-decomp, one bounded step)#2704
Merged
AceHack merged 5 commits intoMay 11, 2026
Conversation
…dence Claude.ai corrected on Aaron's correction: 4 agents stable 3 weeks vs frontier hours baseline. Metrics to measure: PR merge rate, days continuous, drift catch MTBF, gate pass rate. Stability validates methodology, not specific technical claims. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add the required GOVERNANCE §33 archive boundary headers for the forwarded Claude.ai exchange preserved in PR #2698. Co-Authored-By: Codex <noreply@openai.com>
Make the Claude.ai stability archive self-navigable by linking and enumerating Amara's five corrections. Co-Authored-By: Codex <noreply@openai.com>
… bounded step) - Created B-0409 (preamble def, S), B-0410 (amara.ts core, M), B-0411 (README+closure+test, S) - Updated B-0118 frontmatter + added decomposition section (dependency graph) - Enforced Rule 0 (TS over bash): no amara.sh created; pure .ts path - Focused check outcome: ls tools/peer-call/ shows 12 *.ts (incl. amara.ts) + 0 new .sh; rg clean Co-Authored-By: Grok <noreply@x.ai> Co-authored-by: Cursor <cursoragent@cursor.com>
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
There was a problem hiding this comment.
Pull request overview
This PR updates the backlog to re-decompose B-0118 (Amara peer-call headless CLI) into three dependency-ordered atomic child rows, and adds a research archive note capturing an external forwarded exchange about multi-week agent stability.
Changes:
- Added three new P2 backlog rows (B-0409..B-0411) representing the smallest dependency-ordered children for B-0118.
- Updated B-0118 to act as a decomposition index row with child pointers and TS-first framing.
- Added a new docs/research note archiving the forwarded stability/correction exchange and linking to existing Amara corrections memory.
Reviewed changes
Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Description |
|---|---|
| docs/research/2026-05-11-claudeai-three-week-stability-correction-beacon-metrics.md | Adds a dated research archive note with preserved exchange + metrics candidates + link to existing Amara corrections. |
| docs/backlog/P2/B-0409-amara-persona-bootstrap-preamble-definition-ts-first-riven-2026-05-11.md | New atomic child row for defining Amara bootstrap preamble + signature model. |
| docs/backlog/P2/B-0410-amara-ts-core-openai-api-invoke-flag-parity-ts-first-riven-2026-05-11.md | New atomic child row for implementing amara.ts core invocation + flag parity. |
| docs/backlog/P2/B-0411-amara-ts-readme-update-courier-debt-closure-test-invoke-ts-first-riven-2026-05-11.md | New atomic child row for README integration, closure, and a focused invocation test. |
| docs/backlog/P2/B-0118-amara-peer-call-headless-cli-bootstrap-end-courier-debt-2026-04-30.md | Converts parent row into a decomposition index and updates title/status framing. |
Comments suppressed due to low confidence (1)
docs/backlog/P2/B-0118-amara-peer-call-headless-cli-bootstrap-end-courier-debt-2026-04-30.md:10
- P1:
status: decomposedis not a valid backlog status pertools/backlog/README.md(allowed:open/closed/superseded-by-B-NNNN/deferred). This will also make the row inconsistent with its own text later (“status remains open until children land”). Suggest keepingstatus: open(ordeferredif you truly want it off the pickup queue) and recording the decomposition in the body.
Also, last_updated should be bumped to 2026-05-11 (it’s still 2026-05-02) since this file was edited, and the title now says amara.ts while the H1 still references amara.sh—please align the in-body header/wording with the TS-first re-decomp.
status: decomposed
title: tools/peer-call/amara.ts — autonomous bootstrap + communication for Amara (ChatGPT) to end Aaron-courier silent debt (Aaron 2026-04-30; TS-first re-decomp)
tier: factory-tooling
effort: L
ask: Every Amara review this session has been Aaron's manual courier work. The peer-call infrastructure has codex.sh / gemini.sh / grok.sh but no amara.sh; ChatGPT lacks the headless CLI surface that maps to the existing peer-call shape. Until Otto can autonomously bootstrap Amara + do the communication directly, peer-AI review cadence is courier-dependent and incurs silent debt on Aaron. Aaron 2026-04-30 explicitly named this as a constraint Otto must honor.
created: 2026-04-30
last_updated: 2026-05-02
AceHack
deleted the
claim/b0118-decompose-amara-peer-call-ts-first-smallest-atomic-children-riven-2026-05-11
branch
This was referenced May 11, 2026
AceHack
added a commit
that referenced
this pull request
May 14, 2026
…s → B-0457-B-0458 (#3069) * fix(backlog): resolve B-0410-B-0411 collisions — renumber amara series → B-0457-B-0458 Fourth per-collision cleanup from the B-0451 sweep. B-0410 + B-0411 form one connected component (amara series: B-0411 depends on B-0410 within the chain). ## The collisions | ID | B-0118 amara series (2026-05-11 PR #2704) | B-0120 peer-call series (2026-05-11 PR #2706) | |---|---|---| | B-0410 | amara.ts core OpenAI API invoke | peer-call-persona-loader-ts-module | | B-0411 | amara.ts README + courier-debt | grok-ts-persona-flag-impl | ## Resolution: keep B-0120 (peer-call) series Per external-references rule: - B-0120 parent's frontmatter lists `children: [B-0410, B-0411]` AND `depends_on: [B-0410, B-0411]` — strong references - B-0412 (`codex-gemini-ts-persona-flag-impl`) `depends_on: [B-0411]` with body "After B-0411 lands, repeat ... in codex.ts and gemini.ts" — clearly refers to the peer-call grok-ts row - B-0118 parent only has body-text mentions (editable in this PR) → Keep B-0120's children at B-0410-B-0411. Renumber amara series as a unit: B-0410 → B-0457 (amara.ts core) B-0411 → B-0458 (amara.ts README + closure) Internal chain remap: B-0458's `depends_on: [B-0410]` → `[B-0457]`. B-0118 parent body updated to point at new IDs + add note about the renumber. `renumbered_from` + reason recorded in each frontmatter. ## Empirical effect Against current main (which has only B-0409 + B-0410 + B-0411 as remaining collisions after this session's prior 4 cleanup PRs): Duplicate-ID groups: 3 → 1 B-0451 cleanup progress: 9/12 → 11/12. Only B-0409 3-way remains for future-Otto. Co-Authored-By: Claude <noreply@anthropic.com> * shard(tick): 0204Z — fourth B-0451 cleanup; only B-0409 3-way remains Records: PR #3069 opened with B-0410-B-0411 amara series renumber to B-0457-B-0458. Substrate-honest detour from previous tick's restraint (close-out shard): the cleanup is bounded (~10 min following the procedure), the audit tool confirms 3 → 1 group reduction, and the B-0409 3-way remaining is pick-up-able via the procedure memo. B-0451 cleanup progress: 9/12 → 11/12. Only B-0409 3-way remains. Co-Authored-By: Claude <noreply@anthropic.com> --------- Co-authored-by: Claude <noreply@anthropic.com>
AceHack
added a commit
that referenced
this pull request
May 14, 2026
…rate-hygiene sweep (#3073) * fix(backlog): resolve B-0409 3-way collision — completes B-0451 substrate-hygiene sweep Fifth and FINAL per-collision cleanup from the B-0451 sweep. Three rows shared id: B-0409: | Row | Filed | Scope | |---|---|---| | P1 wallet-immune-system | 2026-05-11 10:48 PR #2709 | Wallet immune system spec (L-effort) | | P2 amara-persona-bootstrap | 2026-05-11 10:34 PR #2704 | B-0118 amara series atomic child | | P2 peer-call-ts-audit | 2026-05-11 10:58 PR #2706 | B-0120 peer-call series atomic child | ## Resolution: keep peer-call series at B-0409 Per external-references rule: - B-0120 frontmatter has `children: [B-0409, B-0410, B-0411, ...]` AND `depends_on: [B-0409, B-0410, ...]` — strongest references - B-0118 has body-text mentions only (editable in this PR) - Wallet-immune row has no incoming refs from other rows (composes_with B-0294/B-0321 point FROM the wallet row, not TO it) → Keep peer-call B-0409. Renumber: amara B-0409 → B-0459 (completes the amara series renumber started in PR #3069: B-0410 → B-0457, B-0411 → B-0458, and now B-0409 → B-0459) wallet-immune-system B-0409 → B-0460 ## Chain remap PR #3069 left B-0457's `depends_on: [B-0409]` pointing at the soon-to-be-renumbered amara B-0409. This PR remaps it: B-0457.depends_on: [B-0409] → [B-0459] B-0457.composes_with: ..., B-0409, ... → ..., B-0459, ... Plus B-0118 parent body §Decomposition updated: "B-0409 — Amara persona bootstrap definition" → "B-0459 (renumbered from B-0409) — Amara persona bootstrap definition" `docs/BACKLOG.md` regenerated. ## Empirical effect (CASCADE COMPLETE) ``` $ bun tools/bg/audit-duplicate-row-ids.ts audit-duplicate-row-ids: 561 rows with id field, no duplicate IDs ``` **Down from 12 duplicate-ID groups at session start to 0 on main.** B-0451 cleanup progress: 11/12 → **12/12** (sweep complete). ## Session-arc cascade rounds Round 1: B-0444 P1+P2 (PR #3053) — 12 → 11 groups Round 2: B-0068.1 (PR #3057) — 11 → 10 groups Round 3: B-0090.1-4 batch (PR #3058) — 10 → 6 groups Round 4: B-0370-0373 P2 batch (PR #3065) — 6 → 3 groups Round 5: B-0410-B-0411 amara batch (PR #3069) — 3 → 1 group **Round 6: B-0409 3-way (THIS PR) — 1 → 0 groups** Co-Authored-By: Claude <noreply@anthropic.com> * shard(tick): 0223Z — B-0451 substrate-hygiene cascade COMPLETE (12 → 0 groups) Records: PR #3069 merged (ed9284d). Cross-PR sequencing constraint cleared. Executed the final B-0409 3-way cleanup in PR #3073. audit-duplicate-row-ids on the branch reports: 561 rows with id field, NO DUPLICATE IDs. The substrate-hygiene cascade reaches its terminal state once PR #3073 lands. Session-arc final metrics: - 14 of my PRs merged in the substrate-hygiene cascade - 11 sibling-agent PRs merged (Otto-Desktop, Lior, Codex) - 1 PR in flight (#3073), 1 closed-with-provenance (#3052) - Duplicate-ID groups: 12 → 0 - Dangling-dep refs: 9 → 0 - 12 markdown findings caught across 4 rule classes - 4 Copilot review rounds metabolized on PR #3066 (11 threads) - New tooling: audit-duplicate-row-ids.ts - New durable substrate: procedure memo + B-0451 + B-0456 The work + procedure are durable; future-Otto can pick up B-0456 mechanization or any new findings via the canonical reference. Co-Authored-By: Claude <noreply@anthropic.com> * fix(backlog): bump renumber to B-0462/B-0463 (avoid PR #3070's B-0459-B-0461 reservation) + fix B-0457 body refs Three issues addressed in one commit: 1. **PR #3070 just merged** (2026-05-14) and reserved B-0459/0460/0461 for B-0449 slice 5+ follow-up work (Otto-Desktop's B-0449 body text now points at those numbers). My initial renumber claimed B-0459 (amara) + B-0460 (wallet-immune) — same collision pattern that started this whole sweep, recurring at meta level. Bumped to next-available: B-0462 (amara) + B-0463 (wallet-immune). 2. **Codex P2 + Copilot threads on B-0457**: I updated B-0457's frontmatter `depends_on: [B-0409] → [B-0459]` but the body text still referenced `B-0409` for the preamble source ("Use preamble from B-0409"). After this PR's renumber, B-0409 belongs to a different peer-call-ts-audit row, so the instruction would point at the wrong artifact. Updated B-0457 body to "Use preamble from B-0462 (renumbered from B-0409 per B-0451 sweep)" — same in the Evidence section. 3. **B-0118 parent body**: Updated 3 places from B-0459 → B-0462 to match the new renumber target. ## Final state ``` $ bun tools/bg/audit-duplicate-row-ids.ts audit-duplicate-row-ids: 561 rows with id field, no duplicate IDs ``` The cascade still terminates at zero collisions; just with B-0462 + B-0463 instead of B-0459 + B-0460. ## Substrate-honest meta-observation Otto-Desktop's PR #3070 + my PR #3073 ran in parallel and BOTH reached for the same next-free range (B-0459-0461). The audit tool prevents file-level collisions (same id field in two rows) but doesn't prevent reservation-level collisions (one PR's body text says "we plan to use X" while another PR makes X canonical for different content). This is the next failure class — captured in this commit message for B-0456 follow-up consideration. Co-Authored-By: Claude <noreply@anthropic.com> * shard(tick): 0238Z — caught parallel-renumber-collision with PR #3070; bumped to B-0462/0463 Records: PR #3070 (Otto-Desktop's B-0449 slice-5 reservation renumber) merged, reserving B-0459/0460/0461. My PR #3073 was reaching for B-0459/B-0460 — same collision pattern recurring at meta level. Bumped to B-0462/B-0463. Also resolved 2 Codex/Copilot threads on B-0457 body refs (real bugs — frontmatter updated but body text still pointed at B-0409). Meta-observation captured: audit tool catches file-level collisions but doesn't catch reservation-level collisions where one PR's body text says "plan to use X" while another PR makes X canonical for different content. Next failure class for B-0456 follow-up consideration. Co-Authored-By: Claude <noreply@anthropic.com> * fix(shard): 0223Z — annotate superseded IDs (Copilot+Codex round-1 consistency catch) Three Copilot/Codex threads converged on the same finding: the 0223Z tick shard described the renumber as B-0459/B-0460 (the initial plan) but the PR's final renumber bumped to B-0462/B-0463 (per 0238Z's catch of PR #3070's reservation collision). Without this annotation, future readers following the 0223Z trail would land on the wrong IDs. Added SUPERSEDED annotations inline + a note pointing at 0238Z for the bump narrative. The PR body has also been updated to reflect the final IDs. Cumulative threads resolved across this PR: 3 (Copilot round-1). Co-Authored-By: Claude <noreply@anthropic.com> * shard(tick): 0303Z — PR #3073 round-1 review: 3 cross-shard consistency findings Records: Copilot/Codex caught 3 consistency findings on PR #3073. Same finding-class as PR #3066's round-4 (cross-document drift after mid-PR correction). Fixed in 6006bab: 0223Z shard annotated with SUPERSEDED + pointer to 0238Z; PR body updated to final IDs. This shard was pre-validated by check-md032-blanks-around-lists.ts (the B-0456 helper from PR #3075) before push: "1 file(s) scanned, no MD032 findings". Co-Authored-By: Claude <noreply@anthropic.com> --------- Co-authored-by: Claude <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
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
Decomposed B-0118 (L-effort amara peer-call) into 3 smallest dependency-ordered atomic children per "always re-decompose / assume mistakes" rule + TS over bash (Rule 0). No .sh created; all TS implementation path. One bounded step only; root checkout untouched (dedicated worktree + pushed claim branch used).
Children:
B-0118 updated to
decomposedstatus with child index + graph.Why
Original row too broad (hybrid API choice, L effort, bash+ts mention). Re-decomp splits to atomic, TS-first, dependency-ordered so future loops can pick S/M slices safely. Aligns with TS/Bun migration trajectory (peer-call cluster complete, bash retirement).
Focused checks (included per rule)
ls tools/peer-call/ && rg ...: 12 *.ts present (amara.ts, codex.ts etc already), 0 amara.sh added. Clean.Test plan
Co-Authored-By: Grok noreply@x.ai
Made with Cursor