sentry-selfhosted-mcp

What is sentry-selfhosted-mcp?

A Model Context Protocol (MCP) server designed for interacting with self-hosted Sentry instances. It provides tools to retrieve information and perform actions on issues within a self-hosted Sentry setup, using the instance URL, authentication token, and organization slug from environment variables.

How to use sentry-selfhosted-mcp?

Clone the repository, install dependencies with npm install, build with npm run build, then configure the MCP client to run the built index.js with the required environment variables (SENTRY_URL, SENTRY_AUTH_TOKEN, SENTRY_ORG_SLUG). The server then connects automatically via stdio transport.

Key features of sentry-selfhosted-mcp

• Retrieve details for a specific Sentry issue by ID or URL.
• List all projects within the configured Sentry organization.
• List issues for a project with optional query and status filters.
• Retrieve event details for a specific event ID within a project.
• Update the status of a Sentry issue (resolved, ignored, unresolved).
• Add a comment to a Sentry issue.

Use cases of sentry-selfhosted-mcp

• Debugging and managing errors in a self-hosted Sentry instance from an AI assistant.
• Automating issue triage by fetching and updating issue statuses.
• Adding context or notes to issues via comments during incident response.
• Listing projects and issues to quickly assess the health of a Sentry organization.

FAQ from sentry-selfhosted-mcp

What does sentry-selfhosted-mcp do vs the official Sentry MCP server?

This server is specifically designed for self-hosted Sentry instances, whereas the official server may target Sentry SaaS. It reads connection details from environment variables and requires no external service beyond your own Sentry URL.

What dependencies or runtime are required?

Node.js and npm are required to install dependencies and build the TypeScript code. The server runs as a Node.js process.

How are the Sentry instance URL, token, and organization configured?

They are set via the environment variables SENTRY_URL, SENTRY_AUTH_TOKEN, and SENTRY_ORG_SLUG before launching the server. The token must have scopes: issue:read, project:read, event:read, issue:write, comment:write.

Where does the data live?

All data is fetched from and actions are performed on the self-hosted Sentry instance specified by SENTRY_URL. No data is stored outside this instance.

What transport and authentication does the server use?

The server uses stdio transport for communication with the MCP client. Authentication is via a pre-configured API token passed as an environment variable.