Debugging errors is faster when your AI assistant has full context about what's happening in your application. We just released honeybadger-mcp-server, a new Model Context Protocol (MCP) server that provides AI tools—such as Claude, Cursor, and Copilot—with direct access to your Honeybadger error data and project information.
Automatically fixing a Honeybadger error in Cursor
The MCP server automatically connects your development environment to Honeybadger's API, enabling your AI assistant to manage your Honeybadger projects, retrieve detailed error information and affected user data, access occurrence counts and project reports, and query specific errors with their full context for real-time debugging.
How it works
The MCP server acts as a bridge between Honeybadger's API and MCP-compatible AI tools. Instead of manually copying error details or switching between tools, your AI assistant can directly fetch relevant error information, analyze patterns, and provide contextual debugging suggestions—all within your existing workflow.
Getting started
Pull the Docker image:
docker pull ghcr.io/honeybadger-io/honeybadger-mcp-server:latestGet your personal auth token from your Honeybadger User settings under the "Authentication" tab
Configure your dev environment to run the MCP server (supports Claude Desktop, Cursor, Windsurf, VS Code, Zed, and more)
Start debugging faster with your Honeybadger-aware code assistant
Beta release and roadmap
This is a beta release focused on error tracking and project management—but even with those two features, honeybadger-mcp-server is surprisingly capable. We're actively developing additional tools for working with Honeybadger Insights data, account and team management, uptime monitoring, and other platform features. More to come!
Check out the honeybadger-mcp-server docs for detailed setup instructions for your specific development environment.