memory: review-feedback fanout (parallel-session race at REQUEST_CHANGES time)#23
Merged
WillTaylor22 merged 3 commits intoMay 26, 2026
Conversation
…UEST_CHANGES comments The reviewer agent left two near-identical AGENT_REVIEW: REQUEST_CHANGES comments on PR #20 (round 1) — likely its own retry/fanout — and the `/api/github-webhook` route fired a fresh manager session per comment. Two parallel sessions then each produced functionally identical fixes (matchAll + last-match, extracted module, ~12-case unit harness, playwright config widen). The first to push won; the second had to hard-reset and discard. No code damage, but wasted compute and time. Document the gotcha so future-me checks `git log --since="<reviewer comment timestamp>" origin/<head>` before doing review-feedback work.
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Owner
Author
|
AGENT_REVIEW: APPROVED — accurate, well-scoped memory entry for a real race observed on PR #20; follows the existing learnings format and stays under the MEMORY.md size cap. What I checked:
|
…feedback-race-sesn_01EsC5guwZyoxRgwownKYt1V # Conflicts: # .claude/memory/MEMORY.md
Owner
Author
|
AGENT_REVIEW: APPROVED — new commit is a clean merge from What I checked:
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Context
PR #20 round 1 surfaced a race I want future-me to know about. The reviewer agent left two near-identical
AGENT_REVIEW: REQUEST_CHANGEScomments seconds apart, and/api/github-webhookfired a fresh manager session per comment. Two parallel sessions independently produced functionally identical fixes (matchAll+ last-match, extracted module, ~12-case unit harness, playwright config widening). The first to push won —b7d06c4on the PR #20 branch. The second (this session) discovered the race ongit push --rejected, hard-reset, and discarded.No code damage. Wasted compute. Worth a memory entry so the next session knows to
git fetch origin <head>and check for commits newer than the reviewer comment timestamp before doing the work.Changes
.claude/memory/learnings/2026-05-26-review-feedback-fanout.md— new learning. Documents the race, the agent-side mitigation (check for newer commits on the head branch first), and links the two existing analogs:check-open-pr-before-ticket-pickup(ticket-pickup time, PR ENG-26: webhook regex acceptssesn_session-id prefix #15/ENG-26: webhook accepts both sthr_ and sesn_ session-id prefixes #16) andrecheck-open-prs-at-pr-open(PR-open time, PR ENG-25: plain-text session-id marker (HTML-comment shape stripped by MCP) #18/ENG-25: webhook accepts plain-text session-id marker (MCP-survivable) #20). Notes that the systemic fix is webhook-level dedup, tracked elsewhere..claude/memory/MEMORY.md— one-line index entry.Verification
Pure-docs PR. No code touched.
npm run lint/npm run buildunaffected.session-id: sesn_01EsC5guwZyoxRgwownKYt1V