Datadog Logs

You run a dedicated MCP server that lets you search Datadog logs over HTTP, with configurable queries and time ranges. It exposes an HTTP interface so your MCP clients can request log data and stream updates efficiently.

How to use

Connect to the Datadog Logs MCP Server from your MCP client using the HTTP endpoint exposed by the server. The core MCP tool is designed to search logs with a query, a start time, an end time, and an optional limit on the number of results. Use the SSE connection for MCP communication and POST messages to send commands or queries to the server.

Key capabilities you can leverage through your MCP client include: • Running log searches with a customizable query • Specifying a time window for the search • Controlling how many results are returned in a single response • Receiving updates through a server-sent events channel at the /sse endpoint.

How to install

Prerequisites you need before starting are Docker and Docker Compose, plus valid Datadog credentials (API key and Application key). You will run the server in a container or locally with a development flow.

Install and run with Docker Compose (preferred for quick setup):

docker-compose up

The server will be available at http://localhost:5000

To run in detached mode:

docker-compose up -d

To stop the server:

docker-compose down

Additional setup and development

If you want to run locally without Docker, you can build and start using your Node environment. The following steps are a typical flow for local development.

Install dependencies, build, and start the server locally:

npm install
npm run build
npm start

Available tools

search_logs

MCP tool to search Datadog logs with a query, a start time, an end time, and an optional limit on results.