Skip to content

xpepper/pr-review-agent-skill / copilot-review-loop

copilot-review-loop

Verified

Use when you want GitHub Copilot to review a PR and automatically address its feedback within the current agent session. For any reviewer (human or bot), use pr-review-loop; when an external shell loop should drive one session per comment, use ralph-wiggum-loop.

FileKindSize
README.mdmarkdown1.5 KB
SKILL.mdskill md6.9 KB
references/gh-copilot-review-guide.mdmarkdown1.7 KB
# Copilot Review Loop

An [Agent Skills](https://agentskills.io) skill that automates iterative
GitHub Copilot review loops within a single agent session: trigger Copilot
review, address its feedback one comment at a time, repeat up to 2 cycles.

## What it does

1. Discovers project safeguards from project conventions and runs them
2. Triggers a GitHub Copilot review on the PR
3. Waits for Copilot to complete, then collects its unresolved comments
4. Addresses each comment one at a time: triage → test → fix → test → commit → reply → resolve
5. Re-triggers Copilot review and repeats (max 2 cycles)
6. Posts a final PR summary

If the `pr-review-loop` skill is also installed, delegates the inner loop to it.

## Key property: resumable

Can be interrupted and restarted in a fresh context window at any point without losing progress.

## Prerequisites

- `gh` CLI (required)
- `gh-copilot-review` extension (recommended):
  ```bash
  gh extension install ChrisCarini/gh-copilot-review
  ```
- `pr-review-loop` skill (optional — enhances inner loop)
- PR branch checked out locally

## Install

```bash
npx skills add xpepper/pr-review-agent-skill/copilot-review-loop
```

## Usage

Once installed, describe what you want — the agent activates the skill automatically:

```
Request a Copilot review on this PR and address the feedback
```
```
Trigger a GitHub Copilot review and fix the issues it finds
```

You can also invoke it explicitly:

```
Run copilot-review-loop on PR #42
```
```
Use copilot-review-loop to get and address Copilot's review comments
```
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall