GitHub Sentry

You run an AI-powered MCP server that analyzes code repositories, spots security issues, reviews code quality, and proposes fixes by leveraging GitHub and Sentry integrations. This server helps you accelerate code reviews and remediation by turning error logs and recent commits into actionable improvements.

How to use

You connect a client to the MCP server to receive automated code review insights and suggested fixes. The server fetches recent commits from your repositories, analyzes code quality and security signals, retrieves error logs from Sentry, and returns actionable guidance. Use the MCP client to request reviews on specific projects or pull requests, and review the AI-generated recommendations before applying changes.

How to install

Prerequisites include Python 3.9 or higher, Git, and a preferred virtual environment.

Install Node.js as required for MCP tooling and dependencies.

Step by step commands to set up locally:

# 1. Clone the repository
git clone <repository-url>
cd <repository-folder>

# 2. Install Python dependencies
pip install -r requirements.txt

# 3. Create and configure environment variables
# Create a .env file in the root with the following contents:
AZURE_OPENAI_ENDPOINT="your_azure_openai_endpoint"
AZURE_OPENAI_API_VERSION="your_azure_openai_api_version"
AZURE_OPENAI_API_KEY="your_azure_openai_api_key"
GITHUB_PERSONAL_ACCESS_TOKEN="YOUR_GITHUB_TOKEN"
SENTRY_AUTH_TOKEN="YOUR_SENTRY_TOKEN"

Configuration and startup notes

The MCP server is configured to connect to GitHub for recent commits and to Sentry for error logs. Use the following MCP configuration to enable these two servers.

"mcpServers": {
  "github": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-github"],
    "env": {
      "GITHUB_PERSONAL_ACCESS_TOKEN": "YOUR_GITHUB_TOKEN"
    }
  },
  "sentry": {
    "command": "python",
    "args": ["-m", "mcp_server_sentry", "--auth-token", "YOUR_SENTRY_TOKEN"]
  }
}

Security and access

Keep your tokens and endpoints secret. Use environment variables to configure tokens and access controls, and rotate credentials regularly. Do not hard-code tokens in code or configuration that is committed to version control.

Troubleshooting and notes

If you encounter connection issues, verify that your tokens are valid and that network access to GitHub and Sentry is not blocked. Check that the MCP server processes start without errors and that the required ports are accessible on localhost.

Available tools

code_review

Performs automated review of code for quality and potential issues, highlighting patterns that lead to bugs or security vulnerabilities.

security_vuln_detection

Analyzes code paths and dependencies to detect security vulnerabilities and suggests mitigations.

sentry_error_analysis

Fetches and analyzes Sentry error logs to identify recurring issues and root causes.

issue_fix_suggestion

Proposes concrete fixes and refactors based on analysis results and error context.