CtrlK
BlogDocsLog inGet started
Tessl Logo

handle-pr-comments

Triage and resolve GitHub PR review comments one by one, interactively. Use when the user asks to handle, address, respond to, or resolve PR review comments, or mentions reviewer feedback on a pull request.

76

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Handle PR Comments

Evaluate review comments on the PR (from the current branch, or a PR number/URL if given). For each unresolved comment: summarize the feedback, suggest a resolution, ask the user how to resolve it, apply their choice, commit if needed, and resolve the thread.

  1. Find the PR. Use the given number/URL, else derive it from the current branch (gh pr view). Stop if no PR exists.

  2. Fetch unresolved review threads via the GraphQL reviewThreads field on the pull request — request each thread's id, isResolved, isOutdated, path, line, and comments. Filter to isResolved == false. (Use GraphQL, not REST: only it exposes thread ids and resolution state.)

  3. Loop one comment at a time. For each: (a) summarize the feedback + file/line, reading surrounding code; (b) suggest a concrete fix; (c) ask via AskQuestion with options — apply suggested fix / apply a different fix / reply only / skip; (d) apply the choice; (e) if applicable, summarize what was changed and ask whether to commit; (f) resolve the thread via the GraphQL resolveReviewThread mutation.

  4. Report an overview of what was changed, committed, replied to, skipped, and resolved. Include links.

Notes

  • Handle real-user comments before addressing AI agents' (Copilot, CodeRabbit).
  • Group duplicate comments; when the same/similar issue is mentioned multiple times, handle them all at once.
  • Make a separate commit for each change.
Repository
storybookjs/storybook
Last updated
First committed

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.