Pull Request Review Assistant
Turn any AI chat into a pull request review assistant with this community persona prompt.
by GitHub
GitHub's official MCP server lets an AI agent read and act on your repositories: browse code, open and comment on issues and pull requests, review diffs, and inspect Actions runs. It's offered both as a remote hosted endpoint and as a local server, and authenticates with a GitHub token or OAuth.
The Model Context Protocol (MCP).
Setup steps with copy-paste config. Sources are cited under each method.
Requires: An MCP host that supports remote servers · A GitHub account
Add the remote endpoint
Point your host at GitHub's hosted MCP endpoint and authorize via OAuth — no local install.
https://api.githubcopilot.com/mcp/Authorize
Complete the OAuth flow when prompted; scope the grant to only the repos you want the agent to touch.
Requires: Docker · A GitHub personal access token
Add to your host's MCP config
For Claude Code, add it to `.mcp.json`; other hosts use an equivalent MCP settings block.
{
"mcpServers": {
"github": {
"command": "docker",
"args": ["run", "-i", "--rm", "-e", "GITHUB_PERSONAL_ACCESS_TOKEN", "ghcr.io/github/github-mcp-server"],
"env": { "GITHUB_PERSONAL_ACCESS_TOKEN": "<your-token>" }
}
}
}Restart the host
Reload the agent so it discovers the new server's tools.
Copy-paste prompts written to exercise this plugin, with what to expect.
Your repo has a pile of unlabeled issues.
List open issues in owner/repo with no labels, group them by theme, and propose a label and priority for each. Don't apply anything yet.
What to expect: The agent uses the GitHub tools to pull real issues and returns a grouped, prioritized plan you can approve before it writes labels.
You've pushed a feature branch and want a PR.
Open a pull request from feature/pricing into main. Summarize the diff, list test coverage, and flag anything risky.
What to expect: A PR is created with a diff-grounded description; risky areas are called out rather than glossed over.
Real prompts from the IterPrompt library you can copy or run.
Turn any AI chat into a pull request review assistant with this community persona prompt.
Turn any AI chat into a claude code skill (slash command): push-and-pull-request.md with this community persona prompt.
Review a PR like production depends on it. Purpose-built for typescript & react contexts.
Review a PR like production depends on it. Purpose-built for python backend contexts.
Reviewable pull request descriptions generated from the actual diff.
Turn any AI chat into a github expert with this community persona prompt.
The remote OAuth endpoint is the fastest to set up and needs no Docker. The local server gives you full control over the token and network, which some orgs require.
Yes. It's an MCP server, so it works with Cursor, Windsurf, Cline, VS Code, Zed, and any other MCP-compatible host.
The facts on this page and where they come from. Anything unverified is labeled — we'd rather say "unknown" than guess.
Install steps, pricing, and availability change often. Always confirm on the official page before relying on setup details.