GitHub MCP Server
Official GitHub MCP server — give your AI agent full read/write access to repos, issues, PRs, and actions.
GitHub MCP Server
The GitHub MCP Server is the official Model Context Protocol server from GitHub that exposes the GitHub API as MCP tools. Any MCP-compatible agent (Claude, Cursor, Copilot, GPTScript) can use it to read and write repos, manage issues and pull requests, search code, and trigger Actions — without custom API integration.
Key Features
- Full API coverage: 48 tools across repositories, files, issues, PRs, branches, commits, releases, Actions, and search
- Dynamic toolsets: activate only the tools you need to minimise context window usage
- Safe by default: read-only operations available without write scopes; destructive actions require explicit PAT permissions
- Docker or npx: zero-install
npxmode or a Docker image for server deployments - GitHub Enterprise: works with both github.com and self-hosted GHES instances
- Streaming responses: large file contents and search results streamed efficiently
Quick Start (Claude Desktop)
Add to ~/.claude/claude_desktop_config.json:
{
"mcpServers": {
"github": {
"command": "npx",
"args": ["-y", "@github/mcp-server"],
"env": { "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_..." }
}
}
}
Or run standalone:
export GITHUB_PERSONAL_ACCESS_TOKEN=ghp_...
npx -y @github/mcp-server --transport stdio
npx ai-supply add github-mcp-server
Curated mirror of the open-source GitHub MCP Server (MIT). Get it from the source.
Compromise signals — malicious or tampered code (leaked secrets, backdoors, a dropped executable) — reduce the score, and known dependency CVEs carry a bounded penalty (they warrant review but never QUARANTINE — update the dependency to clear). Other dangerous-by-capability traits are risk surface, expected for some capabilities. Every finding is mapped to its OWASP control below.
Findings mapped to the OWASP Top 10 for LLM Applications (2025) and the OWASP Machine Learning Security Top 10. Expand any flagged control for the exact findings — compromise reduces the score; expected/risk-surface do not, except a known CVE, which carries a small bounded penalty (high/critical → Review).
The same gate an agent runs before installing (POST /api/v1/trust/github-mcp-server/check). Click a policy:
Consume GitHub MCP Server programmatically. Authenticate with an API key or session — see Authorize an agent.
# Agents: CHECK BEFORE YOU INSTALL (no auth) — score, grade, level, capability manifest
curl https://ai-supply.store/api/v1/trust/github-mcp-server
# Gate against your org policy (returns { pass, violations })
curl -X POST https://ai-supply.store/api/v1/trust/github-mcp-server/check \
-H "Content-Type: application/json" \
-d '{"minGrade":"B","denyPermissions":["shell"],"denyUnknownEgress":true}'
# CLI
npx ai-supply add github-mcp-server
# REST (install → download)
curl -X POST https://ai-supply.store/api/v1/listings/github-mcp-server/install \
-H "Authorization: Bearer $AIM_KEY"
# MCP tool
install_listing({ "slug": "github-mcp-server" })OpenAPI spec →Curated mirror — latest upstream source. See the repository for tagged releases.