codecov
A Codecov Model Context Protocol server
claude mcp add --transport stdio stedwick-codecov-mcp npx -y codecov-mcp-server \ --env GIT_URL="XXX" \ --env CODECOV_API_KEY="XXX"
How to use
This MCP server implements a Codecov integration within the Model Context Protocol. It exposes two main capabilities: a tool named get_commit_coverage_totals that fetches the coverage totals for a specific commit and returns a breakdown by file, and a prompt-based helper named suggest_tests that analyzes a Codecov report to propose tests you should write. To use it, run the server with npx as shown in the example configuration, providing your Codecov API key and the repository Git URL. Once running, the MCP protocol will allow you to execute get_commit_coverage_totals to retrieve coverage data for a given commit and then request suggest_tests to receive tailored test suggestions based on the Codecov report. The server is designed to work with Claude or any AI agent by exposing the required endpoints over stdio and accepting standard input/output for command and response messages.
How to install
Prerequisites:
- Node.js and npm installed on your system
- Access to the internet to fetch the MCP server package from npm
Installation steps:
-
Verify Node.js and npm versions: node -v npm -v
-
Install dependencies and build (typical for MCP servers): npm install npm run build
-
Run the server using npx (as indicated in the README): npx -y codecov-mcp-server --api-key YOUR_CODECOV_API_KEY --git-url YOUR_GIT_URL
-
If you need development tooling (optional): npm run inspector # for debugging with the MCP Inspector
Notes:
- Replace YOUR_CODECOV_API_KEY with your actual Codecov API key and YOUR_GIT_URL with the repository's git URL.
- On Windows, you may need the Windows-specific command variant shown in the README.
Additional notes
Tips and caveats:
- Ensure CODECOV_API_KEY and GIT_URL are provided in the environment when running the MCP server configuration.
- Since MCP servers communicate over stdio, debugging can be challenging; using the MCP Inspector tool can simplify debugging and provide a browser-based UI.
- If you see authentication or permissions issues with Codecov, double-check that the API key has access to the repository and that the repository URL is correct.
- The codecov-mcp-server package should be kept up to date to benefit from bug fixes and new features. Run npm install to refresh dependencies when needed.
- For Windows users, use the Windows-specific cursor/command line shown in the README to launch the server correctly.
Related MCP Servers
any-chat-completions
MCP Server for using any LLM as a Tool
bitbucket
Bitbucket MCP - A Model Context Protocol (MCP) server for integrating with Bitbucket Cloud and Server APIs
time
⏰ Time MCP Server: Giving LLMs Time Awareness Capabilities
unity-editor
An MCP server and client for LLMs to interact with Unity Projects
website-publisher
AI Website builder and publisher MCP. Quickly publish and deploy your AI generated code as real website URL. Support html, css, js, python etc.
xgmem
Global Memory MCP server, that manage all projects data.