Skip to content
Datadog MCP Server

Datadog MCP Server

by Datadog
MIT
Repository Docs
Featured Cloud & Infrastructure
observabilitymonitoringlogsmetricstracesincidentsdatadog

Summary

Datadog's official managed MCP server — query logs, metrics, traces, incidents and dashboards from an AI agent, inside your existing Datadog permissions.

Features

  • Managed endpoint hosted by Datadog — nothing to deploy or update
  • Query logs, metrics, traces, incidents and dashboards as MCP tools
  • Runs inside your existing Datadog permissions and governance controls
  • Regional endpoints (US1, EU and other Datadog sites)
  • HTTP transport, compatible with any standards-compliant MCP client
  • Reference agents built on the Claude Agent SDK included in the repo

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

  1. Open your agent's MCP configuration file.
  2. Merge the snippet below into it.
  3. Restart the agent — the "Datadog MCP Server" tools will be available.
{
  "mcpServers": {
    "datadog": {
      "command": "https://mcp.datadoghq.com/v1/mcp"
    }
  }
}

Description

Debugging with an agent breaks down at the moment it needs to know what production is actually doing. Datadog's MCP server closes that gap: it is a managed, Datadog-hosted endpoint that exposes the observability platform as MCP tools, so an agent can pull the logs around an error, the metric that spiked, the trace for a slow request, or the open incident — and then act on it — without anyone pasting screenshots into a chat window.

Managed, not self-run

There is no server to deploy. You point an MCP client at https://mcp.datadoghq.com/v1/mcp over HTTP and authenticate; Datadog operates the endpoint. Regional sites have their own hosts — EU customers use mcp.datadoghq.eu — and the same substitution applies to every other Datadog site.

Permissions are the existing ones

Access runs inside Datadog's own security and governance controls rather than alongside them, so an agent sees what the authenticating identity is already entitled to see. That is the difference between handing an agent observability access and handing it a standing API key with unclear scope.

What you can do with it

Investigate an alert end to end: correlate a metric anomaly with the logs and traces around it, check whether an incident is already open, and summarise the blast radius. The repository ships worked examples built on the Claude Agent SDK, including a monitoring agent that investigates issues and posts findings to Slack, with Datadog LLM Observability wired in so the agent's own behaviour is instrumented too.

Client support

Works with any standards-compliant MCP client — Claude Code, Claude Desktop, Cursor, VS Code and others — via HTTP transport. Announced generally available in March 2026.

Related MCP Servers

MCP: AWS MCP Server

by Amazon Web Services

New

AWS's managed MCP server: call any AWS API, run sandboxed Python against your account, and load AWS agent skills on demand — under IAM, CloudWatch and CloudTrail.

Cloud & Infrastructure

MCP: gcloud MCP Server

by Google

New

Google's official MCP server for driving the gcloud CLI from natural language, with allowlist and denylist controls over which commands an agent may run.

Cloud & Infrastructure
1 views

MCP: MCP Toolbox for Databases

by Google

New

Google's open-source MCP server for enterprise databases — one binary that fronts AlloyDB, BigQuery, Spanner, Postgres, MySQL, Oracle, MongoDB, Snowflake and more.

Databases & DataCloud & Infrastructure
3 views

MCP: Grafana

by Grafana Labs

New

Official Grafana MCP server: search and edit dashboards, query Prometheus, Loki and Pyroscope, and work with alerts, incidents and on-call from an agent.

Cloud & Infrastructure
1 views
Browse all MCP servers →