home / mcp / datadog mcp server

Datadog MCP Server

Provides access to Datadog data sources via MCP for metrics, logs, traces, dashboards, monitors, incidents, and more.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "brukhabtu-datadog-mcp": {
      "command": "datadog-mcp",
      "args": [],
      "env": {
        "DATADOG_API_KEY": "YOUR_DATADOG_API_KEY",
        "DATADOG_APP_KEY": "YOUR_DATADOG_APP_KEY",
        "DATADOG_BASE_URL": "https://api.datadoghq.com"
      }
    }
  }
}

You can run a Datadog MCP Server to enable AI assistants to interact with Datadog’s observability data—metrics, logs, traces, dashboards, and more—through natural language. This server acts as a bridge between your Datadog account and your AI workflows, letting you ask questions and trigger actions across Datadog sources in a conversational way.

How to use

You interact with the server through an MCP client that speaks a standard, model-centric protocol. Start the server locally or in a container, then connect your MCP client to it. Once connected, you can ask questions like querying time-series metrics, listing dashboards, searching logs, examining incidents, or inspecting service dependencies. You can also request read-only information about usage statistics and monitors. Use clear natural language prompts to fetch data, filter results, and read configurations.

Practical usage patterns include: getting current error rates for a service, retrieving the latest traces for a transaction, listing active monitors, checking dashboard configurations, and reviewing SLOs or incident history. Prioritize read-only questions to stay within the server’s security-first posture by default.

How to install

Prerequisites: make sure you have Python and a compatible runtime, and have access to a Datadog account with proper API and application keys. You can run the MCP server from source or via a container image.

From source, install and run the server locally with development-friendly tooling.

git clone https://github.com/brukhabtu/datadog-mcp.git
cd datadog-mcp
pip install -e .

If you prefer running inside Docker, pull the official image and run it with your environment keys.

docker pull ghcr.io/brukhabtu/datadog-mcp:latest

Configuration and security

Required credentials are your Datadog API key and Datadog App key. These keys enable the MCP server to access your Datadog data. Optional settings control transport, timeouts, and security filtering.

Set the following environment variables to enable the server and tailor its behavior.

DATADOG_API_KEY="your-datadog-api-key"
DATADOG_APP_KEY="your-datadog-application-key"
```

Optional examples to customize behavior include: 
```
DATADOG_BASE_URL="https://api.datadoghq.com"
DATADOG_TIMEOUT=30
MCP_TRANSPORT=stdio
MCP_PORT=8000
MCP_LOG_LEVEL=INFO
MCP_ENABLE_SECURITY_FILTERING=true

Troubleshooting

If authentication fails, verify both keys are correct and have the right permissions for the regions you use. If connection issues occur, ensure network reachability to the Datadog API and adjust timeouts as needed. If you encounter missing specifications, confirm that the OpenAPI spec is accessible at the expected location and that versions align with your Datadog API.

Development

If you want to contribute, set up a development environment, install dependencies, and run tests to verify changes.

Available tools

query_metrics

Query time-series metrics, list metrics, and retrieve metadata for Datadog time-series data.

search_logs

Search and filter log events across your Datadog data.

apm_traces

Access tracing data, service maps, and dependencies for applications.

infrastructure

Retrieve host, container, and process metrics for your environment.

dashboards

List and read dashboard configurations to understand visual monitoring.

monitors

Access monitor rules and their current status.

incidents

Track and manage incidents within the Datadog platform.

service_catalog

Retrieve service definitions and relationships for your environment.

slo

Access service level objectives and related compliance data.

usage

Get account usage statistics and related metrics.