
Copilot PR Autopilot
Loops Copilot code review on a GitHub PR: triggers, triages, fixes, replies and resolves until converged.
- Repository
- github/awesome-copilot
- Stars
- 40,703
- Licence
- MIT
- Files
- 15 files
What it does
Once installed, the agent can take a pull request on GitHub and repeatedly request a Copilot code review, read the comments Copilot leaves, sort them into fix, decline or escalate, make code fixes in parallel, run the repo's build and tests, commit and push, reply to each thread, and resolve the ones it handled. It keeps going until Copilot has nothing new and every thread has a reply, then reports back. Open threads left for a human are treated as deliberate hand-offs rather than failures.
Great for
- Developers who get many Copilot review comments on a PR and want them worked through automatically
- Maintainers with Triage or Write permission on a GitHub repo who want hands-off review iteration
- Teams already using Copilot Code Review and the gh CLI
Skip it if
- PRs still under active design, since findings churn each round and the loop wastes effort (the skill itself says so)
- Users who want human reviewer feedback rather than Copilot's, or who cannot tolerate automated commits and pushes to their branch
- External PR authors without write permission, who only get a single-iteration mode and must manually re-trigger Copilot
Try asking
- “Address the Copilot comments on this PR and push the fixes”
- “Run a Copilot review loop on my current pull request until it converges”
- “Fix this PR using the open Copilot review threads”
- “Iterate on Copilot feedback for PR 123 and tell me which threads you escalated to me”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 15 original files, the MIT licence and an attribution note.
copilot-pr-autopilot.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip copilot-pr-autopilot.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip copilot-pr-autopilot.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Only run it on PRs whose structure is stable; early in design the loop churns through findings without converging
- Check the escalated threads at the end, since the skill hands design tradeoffs and declined items to you rather than resolving them
- Review the commits it pushes before merging, as it commits and pushes automatically on your branch
What's inside
- references/01-request-review.md
- references/02-wait.md
- references/03-list-threads.md
- references/04-triage.md
- references/05-fix.md
- references/06-build-test.md
- references/07-commit-push.md
- references/08-reply-resolve.md
- references/09-convergence.md
- references/10-cleanup.md
- references/api-quirks.md
- references/orchestration.md
- SKILL.md
- templates/reply-decline.md
- templates/reply-drift.md
Redistributed unmodified from github/awesome-copilot under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.