mcp -datadog

This MCP server exposes Datadog API capabilities through the Model Context Protocol to allow programmatic access to Datadog data and actions. It implements a collection of tools that map to common Datadog resources, including incidents, monitors, logs, dashboards, metrics, traces, hosts, and RUM. Clients can call these tools to list, fetch, or search resources, enabling integration with monitoring workflows, incident response, and observability dashboards. The server emphasizes extensibility, so additional Datadog endpoints can be added to support new features over time.

Prerequisites:

• Node.js and npm installed on your system
• Internet access to fetch the MCP server package from npm

Installation steps:

1. Ensure you have Node.js and npm installed
   • Check versions: node -v and npm -v
2. Install the MCP server package globally or run via npx (recommended for this project)
   • npm install -g @winor30/mcp-server-datadog (optional if you prefer a global install)
3. Run the MCP server through your MCP orchestrator or directly with npx
   • npx -y @winor30/mcp-server-datadog
4. Verify the server starts and listen on the expected port (default MCP port) and check logs for readiness

If you use a containerized approach later, you can adapt this to docker accordingly.

Tips:

• This server uses the Datadog API; you may need Datadog API keys or credentials configured in your environment (depending on how the underlying package is implemented). Prepare environment variables such as DD_API_KEY or DD_APP_KEY if required by the Datadog SDK used by the server.
• The Tools section documents available operations. When integrating, ensure you provide the correct inputs (IDs, filters, time ranges) as described per tool.
• If you encounter rate limits or authentication errors, verify Datadog credentials scope and consider implementing retry logic at the consumer layer.
• For local development, you can extend or customize the MCP server to include additional Datadog endpoints by following the existing tool pattern in the repository.