Datadog MCP in Claude Code

Quick answer: Install the Datadog MCP server with npx -y @datadog/mcp-server, add the JSON block below to ~/.claude/settings.json, restart Claude Code, and run /mcp to confirm the connection. Setup runs about 5 minutes on a fresh machine, verified on @datadog/mcp-server as of April 15, 2026.

The Datadog MCP server gives Claude Code a tool surface on the Datadog platform. After setup, the model can query metrics, search logs, inspect monitors, read dashboards, and explore traces. The server is maintained by Datadog as @datadog/mcp-server and uses standard API plus app keys for auth.

This guide covers what you get after the wiring is done, the exact config, verification steps, prompt patterns that tend to work well, and the 4 issues that trip people up most often in the first week.

What you get when it is connected

Once the Datadog server is attached, Claude Code can call the server tools from inside any conversation. You do not invoke the tools by hand. When you ask Claude a question the model decides which tool to call and parses the response for you. For teams that live inside Datadog day to day, this replaces dozens of context switches per week with a single line in chat.

Tools exposed include query_metrics, search_logs, list_monitors, get_monitor, list_dashboards, get_dashboard, search_traces, get_slo. Query volumes scale with your Datadog plan - the MCP server respects the same rate limits as direct API calls, which is generally 300 requests per hour per app key.

Prerequisites

A Datadog account, an API key from app.datadoghq.com/organization-settings/api-keys, and an application key from the same settings area. The site region (US1, US3, EU1) matters since the API endpoints differ. Node 20 or later.

If you use a version manager like nvm or asdf for Node, confirm the version Claude Code inherits. Open a terminal, run node -v, and note the output. Claude Code uses the Node it sees on PATH at launch, so a shell profile that sets the right version is the reliable path.

Install via npx

Run the package once with npx to verify it starts cleanly:

npx -y @datadog/mcp-server

The first run downloads the package (a few MB) and starts the server on stdio. The server does not print much on success - it waits for MCP protocol messages on stdin. Press Ctrl-C to stop it. The actual runtime setup happens through Claude Code itself in the next step.

If the install fails with a network error, your npm registry may be blocked. Set npm config set registry https://registry.npmjs.org and retry. Behind a corporate proxy, also set HTTP_PROXY and HTTPS_PROXY in your shell.

Add the config block to ~/.claude/settings.json

Open ~/.claude/settings.json in your editor. If the file does not exist yet, create it with {} as the starting content. Add an mcpServers object with an entry for this server:

{
  "mcpServers": {
    "datadog": {
      "command": "npx",
      "args": ["-y", "@datadog/mcp-server"],
      "env": {
        "DATADOG_API_KEY": "XXX",
        "DATADOG_APP_KEY": "YYY",
        "DATADOG_SITE": "datadoghq.com"
      }
    }
  }
}

Save the file. If you already have other MCP servers defined, merge the new entry into the existing mcpServers object rather than replacing it.

Restart Claude Code fully (quit and reopen, not just close the window). The server is spawned lazily on the first tool call in a session, not at launch, but the config is read once per Claude Code start.

Verify the connection

Open a new Claude Code session and type /mcp at the prompt. You should see the server listed with a green or connected indicator. If it shows as failed, click into it for the stderr output - the error message usually points at the problem directly (bad token, wrong path, missing Node).

Run a trivial first prompt to confirm round trips work. Good smoke tests:

• For read servers: ask for a list of whatever resource type it exposes.
• For write servers: ask for a describe on a known resource first, then try a safe write on a test resource.

If the first prompt works, the wiring is done. From here on you interact with the server purely through normal prompts in Claude Code.

Example prompts that work well

Here are prompts that tend to get good responses once the server is attached:

• Query the metric system.cpu.user averaged by host over the last hour and tell me the top 5 hosts by CPU usage.
• Search logs for service checkout-api with status error in the last 30 minutes and summarize the error messages.
• List every monitor whose status is Alert and tell me which service each one covers.
• Read the dashboard titled Production Overview and summarize the key widgets.
• Find APM traces in service payments-service with latency over 2 seconds in the last 15 minutes.
• Check the uptime SLO for service api and tell me the current 30-day burn rate.

Claude will chain tool calls on its own when the prompt implies several steps. For a summarize-then-write flow the model will often call read tools first, then a single write tool at the end. If a prompt keeps burning tool calls, narrow it: specify the resource ID, the time range, or the exact field you want rather than asking Claude to scan everything.

Environment variable security

Three env vars: DATADOG_API_KEY for request auth, DATADOG_APP_KEY for read access to dashboards and monitors, and DATADOG_SITE for the region. Create scoped app keys via the Datadog UI to limit what Claude can read. Rotate both keys from the API Keys page if they leak. Never commit either value.

A general rule across every MCP server: never paste secrets directly into settings.json that lives in a shared or git-tracked directory. Keep the actual secret values in your shell profile (~/.zshrc, ~/.bashrc, or a 1Password-cli helper), export them at shell start, and reference the variable names from the Claude config. That way the secret stays on your machine and the config file is safe to share with teammates.

On macOS, terminals launched from Spotlight or from the Dock both inherit the shell profile. If you launch Claude Code from a GUI shortcut that does not go through a shell, env vars may not propagate - launch from a terminal instead.

Troubleshooting

Tool calls return 403 Forbidden. The app key lacks the scope the call needs. Datadog restricted app keys need explicit scopes for metrics:read, logs:read, etc. Edit the key and add the missing scopes, then retry.

Metric queries return no data. The metric name or tag is wrong, or the time window is outside retention. Check the metric exists in the Metrics Explorer and confirm the tag spelling. Default metric retention is 15 months.

Log search returns too few results. Log indexes filter out low-priority logs by default. Check the index configuration at app.datadoghq.com/logs/pipelines/indexes to make sure the logs you want are actually indexed, not just ingested.

Calls fail with site mismatch. The DATADOG_SITE env var must match where the account lives. US1 uses datadoghq.com, EU1 uses datadoghq.eu, US3 uses us3.datadoghq.com. Pick the one shown in your Datadog URL.

For any issue not listed here, the first step is /mcp inside Claude Code to see the current status and any recent stderr from the server. The second step is running the exact npx command from your terminal to see if the server starts cleanly outside Claude Code. Between those two checks, most problems become obvious within a minute.

Next steps

Once the Datadog server is attached and verified, the useful next move is writing a short prompt template you keep in your notes. List the 3 or 4 prompts you run most often against this server, and paste them into Claude Code when needed. Over a few weeks you build a personal command library that gets real work done without typing much.

For team projects, commit a .mcp.json at the repo root with the same structure. Everyone on the team gets the server wired up automatically on first open, and individual secrets stay in shell profiles. That is the setup pattern that scales past a single developer.