Repository navigation
fix: complete native multi-plan lifecycle and setup discovery (Factory-038) - #128
Draft
PatrickSys wants to merge 4 commits into
Draft
PatrickSys wants to merge 4 commits into
PatrickSys wants to merge 4 commits into
Conversation
Agent-Harness: chat-on-steroids Agent-Session: factory-038-fix1
Agent-Harness: codex-cli Agent-Session: factory-038-fix1-revision
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Agent-Harness: codex-cli Agent-Session: factory-038-fix6
Agent-Harness: codex-cli Agent-Session: factory-038-fix6-hardening
This branch was successfully deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem and resulting behavior
Factory-037 installed consumers exposed two lifecycle correctness gaps: PLAN02 failed with stale_state while PLAN01 remained the recorded authority, and an amended verification report left an outdated gaps_found state. Setup also failed to expose a root AGENTS.md workflow map. This stacked draft provides a guarded second-PLAN handoff, accurate per-plan routing/status and safe workflow discovery.
Stacked on PR #127 (factory/036-real-run@d93dd81c). Factory-038 commits are isolated on pushed factory/038@9963bf9. Neither PR was merged to default branch or published to npm.
Delivered source fixes
Final HEAD verification: node tests/run-all.cjs 34/34 test files GREEN. Package from this exact commit built with Node24/npm11, version retained 0.35.0, 122 files, SHA256 1FDE384488B6CBBDFD1602F6BB57E7D54DA693288087D5E95790FD091ACD2995. Fresh source-free install setup --yes exit0 / health --json exit0; 25-line root AGENTS.md and portable work-plan skill present. Exact receipts: %TEMP%/factory-038/logs/final-head/, logs/pack-final-9963/, install-smoke-final-9963/.
Honest installed consumer result (REVISE)
T3 ran earlier integrated commit 7ad1537 package SHA256 98AF33C248807FE0181151F66B2751E1076849BA4D48AB268A7607285197FE91 on fresh angular-spotify@7af8b43. Two application writers overlapped and produced both feature slices; 31.8m, 15.38M top-level tokens, 17 scoped tests and default/production builds GREEN. Native documentation led the agent to ONE brownfield CHANGE with two tasks. It self-approved that one plan (not Patrick), closed it through verify/audit transition, and next correctly said brownfield_change_closed. It did not exercise two independently routed PLAN01 and PLAN02 lifecycles.
Independent native reviewer: REVISE for requested two-plan acceptance; GO only for truthful single-chain closeout. Separate blind app reviewer scored 12/12 feature criteria but REVISE, merge NO, because frozen host comparison caught ten new no-explicit-any warnings in consumer test specs (9 playback.service.spec.ts, 1 track-notification.service.spec.ts), over the existing lint-store circular-dependency baseline error. Host build + eight relevant library tests GREEN, app test/lint app old baseline failures unchanged. Strict collector status REGRESSION is correct; do not waive it. No live Spotify account or native OS notification proof. The consumer findings concern the disposable trial code and do not imply an additional Workspine-source regression.
Source code fixes remain GREEN. The final-HEAD packed smoke is real, but T3 was not repeated against 9963bf9 after hardening. Conditional standalone T2 rerun did not trigger because T3 failed. See PR #127 "Real tasks (038)" for the 037-comparable table, frozen controls, timing and evidence paths.
Evidence, authorization, rollback