You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
test(source-control): add a setup eval case that writes only what the user confirmed #7998
Type: Task Summary: One plugin-eval case shows source-control setup writes only the settings the user confirmed.
Current behavior:
plugins/source-control/evals has no setup case.
Desired behavior:
A case in the layout of the existing source-control cases runs setup in a scratch repository where the prompt confirms one change and declines another; deterministic graders check that only the confirmed change is written. No gh fixture is needed.
Key interfaces:
The plugin-eval case layout.
The setup skill's confirm-before-write contract.
Acceptance criteria:
The case directory exists with case.yaml, prompt.md, scaffold and graders.
A grader fails when the declined change appears in the workspace.
A source-control changelog fragment is added and scripts/check-changelog-fragments.sh passes.
Out of scope:
babysit cases and the gh stand-in.
Paid runs.
This was generated by AI during backlog shaping. Split from #7148; the owner approved filing it in an attended backlog-shaping session on 2026-10-18.
Activity