home / mcp / datadog mcp server

Datadog MCP Server

MCP Server for Datadog monitoring with Kubernetes, APM, and infrastructure metrics support.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "adisaputra10-mcp-datadog": {
      "command": "npx",
      "args": [
        "mcp-datadog-playcourt"
      ],
      "env": {
        "DATADOG_SITE": "datadoghq.com",
        "DATADOG_API_KEY": "your-api-key",
        "DATADOG_APP_KEY": "your-app-key"
      }
    }
  }
}

You deploy and run a Datadog-focused MCP Server that collects Kubernetes, deployment, namespace, host, and APM metrics, enabling centralized monitoring through your MCP client. This server integrates Datadog observability capabilities into your MCP workflow so you can query metrics and monitor your infrastructure directly from your MCP environment.

How to use

Use your MCP client to connect to the Datadog MCP Server you configured. You can run queries to retrieve Kubernetes metrics, monitor deployments, aggregate namespace resource usage, enumerate hosts, and access APM-related insights. The server exposes a set of actions (tools) you can invoke from your MCP client to fetch the exact metrics you need and to drill into details for a specific host or deployment.

How to install

Prerequisites: you need Node.js installed on your machine so you can run npm and npx commands.

Step 1. Install and run the MCP server client command directly with npx.

npx mcp-datadog-playcourt

Additional configuration and usage notes

Configure your MCP client to load the Datadog MCP Server with the required environment variables for Datadog access.

{
  "servers": {
    "datadog": {
      "command": "npx",
      "args": ["mcp-datadog-playcourt"],
      "env": {
        "DATADOG_API_KEY": "your-api-key",
        "DATADOG_APP_KEY": "your-app-key",
        "DATADOG_SITE": "datadoghq.com"
      }
    }
  }
}

Security and environment considerations

Keep your Datadog API and Application keys secure. Do not share these keys in public repositories or logs. Store keys in a secure secrets manager and reference them in your MCP settings as shown.

Available tools

query_kubernetes_metrics

Query Kubernetes metrics such as pod counts, CPU, memory, and network usage.

query_deployment_pods

Retrieve deployment metrics including per-pod metrics and response time for deployments.

get_namespace_metrics

Aggregate resource usage at the namespace level.

list_all_hosts

List all monitored infrastructure hosts with system information.

get_host_detail

Fetch detailed information for a specific host.

list_hosts

List all monitored hosts.

list_databases

List all monitored databases.

query_host_metrics

Query host infrastructure metrics.

query_database_metrics

Query database metrics.