feat: record the model and outcome of every AI round on the PR - #28
Merged
Merged
Conversation
Each round resolves its model once (pick step) and every comment it posts carries an ai-meta marker (run id, round, mode, model); the step that sets the terminal label writes an ai-outcome marker for the same run. The status comment carries earlier runs' markers forward on a retry. Read by the model-tier report in Talieisin/automation.
Copilot started reviewing on behalf of
george-elphick-talieisin
September 21, 2026 22:34
View session
There was a problem hiding this comment.
Copilot review overview
🟡 Changes recommended
Unresolved metadata-schema, failure-outcome, and requeue-recording issues remain.
Get a fresh assessment by requesting another Copilot review.
Review effort: Lite
Findings: 1
Open (3)
What changed in this PR
Adds persistent model and outcome metadata to Dependabot and Copilot-response workflow rounds.
Changes:
- Resolves and records the model used for each AI round.
- Adds hidden metadata markers and visible model attribution.
- Preserves metadata across retries and records terminal outcomes.
| File | Summary |
|---|---|
.github/workflows/dependabot-upgrade.yml |
Records round models and upgrade outcomes. |
.github/workflows/copilot-response.yml |
Records Copilot-response models and CI outcomes. |
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
pickstep that--modelreads), and every comment or review it posts carries a hidden<!-- ai-meta v=1 run=… round=… mode=… model=… -->marker plus a visibleModel:line.<!-- ai-outcome v=1 run=… outcome=… -->:ci-watch, the decide/suggest park paths, suggest success, requeues, thefinalizecrash path, and the Copilot-response CI result. Outcomes are recorded by the run itself because labels change later for unrelated reasons (the Copilot-response CI result, the dispatcher's stale-green check).Test plan
main(only shifted line numbers on existing SC2016 notices)bash -non everyrun:block in both workflowsgh: requeue → push → ci-watch success → retry blocked keeps all markers in order; a gate-blocked run adds none; a suggest park and a no-push Copilot round are both marked correctlyai-metaandai-outcomewith the samerun=, and the model matches--modelin that run's logNeed help on this PR? Tag
@codesmith-botwith what you need. Autofix is disabled.