Repository navigation
Promote checked restart custody and readiness repair - #293
Merged
Merged
Conversation
Fix managed restart custody and readiness adoption
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Outcome
Promote managed restart recipe refresh and manager-owned readiness adoption from checked staging.
Work authority
Work item / Issue: Current cli backlog: retained requirements after issue consolidation #274; SDK #376
Proposal / decision: N/A — human-directed capacity-provider acceptance repair, not a managed agent proposal.
Assignment / checkpoint: N/A — this Codex workspace task; exact owning Issue is checkpoint authority.
Actor: Codex under adrianwebb
Human authority: adrianwebb explicitly authorized fixes, checked merges and temporary branch cleanup.
Agent / capacity provider: Codex desktop; no managed capacity-provider assignment used for this change.
Exact base ref: main 8bece29
Exact head ref: staging 63c7928
Agent-authored under human authority
Plan
Reproduce owning failure, repair with original unit/integration/scene coverage, merge checked staging through ordinary history-preserving PR, then main after automated checks. No tags/releases/new local branches or worktrees. Temporary remote repair branch removed only after ancestry proof.
Changes and commits
5e79883;01a0bbd0c4d6a0291e6a28b5656b355ce3df476f;staging merge63c7928b2ab2456e495a0e55572b9a47d379315c. No architecture, credential, model, allowance, watchdog or UI change.
Verification
Original npm test complete366/366 and165 coded component steps PASS in Actions37946657623; fresh main PR verify37947775006 and staging verify37947709039 PASS. Exact source digests and original receipt independently audited. Initial PR description validation failed because I omitted the repository template; this description corrects metadata only, with failed run37947775642 retained.
Risk and rollback
Readiness recovery affects managed development selections only. On regression, stop new acceptance admission and revert the exact repair commits through a checked PR; never reset protected history, delete held claims or fabricate settlement. Deployment runtime recovery uses the original guarded dev host activation/rollback path, not sudo or arbitrary commands.
Completion summary
Owning repair is committed and staging-merged. This is main promotion only; full SDK acceptance remains OPEN and needs fresh all-owner prerequisites before one new original native campaign. No component totals count as managed golden completion.
Submission checklist