Netdata MCP: ask your AI app what your infrastructure is doing

Per-second metrics, logs, alerts and ML-based anomaly detection from Netdata's real-time infrastructure monitoring can sit inside your AI app's chat, so it can help you work out why a node is struggling. ContextGate connects Netdata's MCP server to Claude, ChatGPT and Cursor with your Netdata API token, and each metric query or alert lookup made during an incident is kept for review.

  • Ready to connect
  • Free

Connect Netdata monitoring to Claude, ChatGPT or Cursor

  1. 1Click Connect Netdata free and paste an API token from your Netdata account.
  2. 2Netdata's tools are listed once the token is accepted, and all of them are on. Keep the ones that read metrics, logs and alerts, and switch off anything in the list you don't want an AI app touching mid-incident.
  3. 3Add ContextGate to your AI app as a connector or MCP server. The steps for each app are on the Works with page.
  4. 4Ask your AI app which Netdata alerts fired on a server in the last hour, then open Activity to see the call behind the answer.

Troubleshoot with Netdata from Claude or Cursor

  • Pull the per-second metrics around a CPU or memory spike into the chat.
  • Read a host's logs next to the error you're debugging in your editor.
  • Check which alerts are active before you decide whether to wake someone up.
  • Ask what Netdata's ML-based anomaly detection flagged on a node.

Monitoring access your workspace can see

  • Switch off any Netdata tool you don't want in use; your AI app no longer sees it, and calls to it are refused.
  • Policies check every call to Netdata, so a secret pasted into a question can be blocked before it leaves.
  • Activity records each metrics and log lookup, so you can see what an AI app looked at during an outage.

What you get with Netdata and ContextGate

  • Your AI apps can use NetdataClaude, Cursor, ChatGPT and other AI apps reach Netdata through the one ContextGate connection you add to them.
  • You choose which tools they can useSwitch off any Netdata tool you don't want used. It's hidden from your AI apps and refused if called.
  • Every call is checked and recordedYour policies check each call (for example, personal data is hidden and secrets are blocked), and Activity keeps a record of it.

ContextGate sees only the tool calls that go through it, not your AI app's built-in tools or your chats.

Netdata tools

Netdata's own MCP server shows its tools once you sign in. When you connect it, they are listed and synced in your workspace.