MCP GitHub CrunchTools
A secure MCP server for interacting with GitHub issues, pull requests, repository files, and search, supporting both github.com and GitHub Enterprise Server.
README
MCP GitHub CrunchTools
A secure MCP (Model Context Protocol) server for GitHub issues, pull requests, repository files, and search. Works with github.com and GitHub Enterprise Server.
Overview
This MCP server is designed to be:
- Secure by default - Comprehensive input validation and token protection
- No third-party services - Runs locally via stdio, your API token never leaves your machine
- Multi-instance - Works with github.com or GitHub Enterprise Server via configurable API URL
- Cross-platform - Works on Linux, macOS, and Windows
- Automatically updated - GitHub Actions monitor for CVEs and update dependencies
- Containerized - Available at
quay.io/crunchtools/mcp-githubbuilt on Hummingbird Python base image
Naming Convention
| Component | Name |
|---|---|
| GitHub repo | crunchtools/mcp-github |
| Container | quay.io/crunchtools/mcp-github |
| Python package (PyPI) | mcp-github-crunchtools |
| CLI command | mcp-github-crunchtools |
| Module import | mcp_github_crunchtools |
Why Hummingbird?
The container image is built on the Hummingbird Python base image from Project Hummingbird, which provides:
- Minimal CVE exposure - Built with a minimal package set, dramatically reducing the attack surface
- Regular updates - Security patches are applied promptly
- Optimized for Python - Pre-configured Python environment
- Production-ready - Proper signal handling and non-root user defaults
Features
Issues (3 tools)
list_issues_tool- List issues for a repository (pull requests excluded)get_issue_tool- Get a single issue by numbercreate_issue_comment_tool- Comment on an issue or pull request (write)
Pull Requests (4 tools)
list_pull_requests_tool- List pull requests for a repositoryget_pull_request_tool- Get a single pull request by numberget_pull_request_diff_tool- Get the unified diff for a pull requestget_pull_request_checks_tool- Combined CI status (check-runs + commit status)
Files (2 tools)
get_file_content_tool- Read decoded file content from a repositorylist_repo_tree_tool- List the git tree (files and directories)
Search (2 tools)
search_code_tool- Search code across GitHubsearch_issues_tool- Search issues and pull requests across GitHub
Installation
With uvx (Recommended)
uvx mcp-github-crunchtools
With pip
pip install mcp-github-crunchtools
With Container
podman run -e GITHUB_TOKEN=your_token \
quay.io/crunchtools/mcp-github
Configuration
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
GITHUB_TOKEN |
Yes | — | GitHub Personal Access Token |
GITHUB_API_URL |
No | https://api.github.com |
API base URL (set for GHES) |
GITHUB_DEFAULT_ORG |
No | — | Default owner when a tool omits owner |
Creating a GitHub Personal Access Token
-
Navigate to token settings
- Go to https://github.com/settings/tokens
-
Create a token
- Name:
mcp-github-crunchtools - Expiration: Set an appropriate date (90 days recommended)
- Scopes: Grant read access to contents, issues, and pull requests.
Add write to issues/PRs only if you need
create_issue_comment_tool.
- Name:
-
Copy and Store Token
- Copy the token immediately (shown only once)
- Store securely in a password manager
Add to Claude Code
claude mcp add mcp-github-crunchtools \
--env GITHUB_TOKEN=your_token_here \
-- uvx mcp-github-crunchtools
For GitHub Enterprise Server:
claude mcp add mcp-github-crunchtools \
--env GITHUB_TOKEN=your_token_here \
--env GITHUB_API_URL=https://ghe.example.com/api/v3 \
-- uvx mcp-github-crunchtools
For the container version:
claude mcp add mcp-github-crunchtools \
--env GITHUB_TOKEN=your_token_here \
-- podman run -i --rm -e GITHUB_TOKEN quay.io/crunchtools/mcp-github
Usage Examples
List Issues
User: List open issues for crunchtools/mcp-github
Assistant: [calls list_issues_tool with owner="crunchtools", repo="mcp-github"]
Review a Pull Request
User: Show me the diff for PR #5 in crunchtools/mcp-github
Assistant: [calls get_pull_request_diff_tool with pull_number=5]
Check CI Status
User: Did the checks pass on pull request 5?
Assistant: [calls get_pull_request_checks_tool with pull_number=5]
Read a File
User: Show me src/server.py from crunchtools/mcp-github
Assistant: [calls get_file_content_tool with path="src/server.py"]
Search
User: Find code using FastMCP in crunchtools repos
Assistant: [calls search_code_tool with query="FastMCP org:crunchtools"]
Security
This server was designed with security as a primary concern. See SECURITY.md for details.
Key Security Features
-
Token Protection
- Stored as SecretStr (never accidentally logged)
- Environment variable only (never in files or args)
- Sanitized from all error messages
-
Input Validation
- Pydantic models for write inputs
- Allowlist character validation for owner/repo names
- Path traversal prevention for file reads
-
API Hardening
- Bearer-token auth and pinned GitHub API version
- HTTPS enforcement (except localhost)
- TLS certificate validation
- Request timeouts (30s)
- Response size limits (10MB)
-
Automated CVE Scanning
- GitHub Actions scan dependencies
- Container security scanning with Trivy
Development
Setup
git clone https://github.com/crunchtools/mcp-github.git
cd mcp-github
uv sync --all-extras
Run Tests
uv run pytest
Lint and Type Check
uv run ruff check src tests
uv run mypy src
Build Container
podman build -t mcp-github .
License
AGPL-3.0-or-later
Contributing
Contributions welcome! Please read SECURITY.md before submitting security-related changes.
Links
<!-- mcp-name: io.github.crunchtools/github -->
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.