| name | apply-review-feedback |
| description | Classify and apply accepted PR/MR review feedback. |
| model | sonnet |
| allowed-tools | ["Read","Write","Edit","Glob","Grep","Bash","WebFetch","mcp__*"] |
Run the review-feedback-applier agent on the current branch's open PR/MR.
Fetches review comments, classifies feedback as accepted/rejected/ambiguous,
and applies accepted changes to local files. No auto-commit or auto-push.
User invocation:
/apply-review-feedback [options]
Examples:
/apply-review-feedback
/apply-review-feedback --pr 42
/apply-review-feedback --gitlab
/apply-review-feedback --mr 15
/apply-review-feedback --github --pr 100
$ARGUMENTS
--github | --gitlab: Force platform detection. OPTIONAL.
--pr | --mr : Explicit PR/MR number. OPTIONAL (auto-detected from branch).
Parse $ARGUMENTS and delegate to the `review-feedback-applier` agent with the parsed options.
The agent handles all platform detection, pre-flight checks, classification, and application.
This command is a thin entry point — all logic lives in the `review-feedback-applier` agent.
Non-interactive: do not depend on follow-up questions; use $ARGUMENTS, safe defaults, or NEEDS_INPUT.
Idempotent: reruns are safe; already-applied changes are detected via code context comparison.
<output_format>
<what_to_return>Classification summary (accepted/rejected/ambiguous counts), list of modified files, artifact paths under tmp/review-feedback/<branchPath>/, and reminder to review and commit manually.</what_to_return>
</output_format>
<user_input>$ARGUMENTS</user_input>