Skip to main content
This section covers using ngrok from an AI agent, such as Claude Code, Codex, or Cursor. For the ngrok agent that runs on your machine and forwards traffic, see ngrok agents.

​
Manage your account with the ngrok MCP server

The ngrok MCP server lets your AI agent read and change your ngrok account. Ask it about your endpoints, traffic, usage, and team, or ask it to create endpoints, write Traffic Policy, and manage IP policies.

​
Teach your AI agent ngrok tasks with skills

ngrok skills teach your AI agent how to use ngrok on your machine. With them, it can expose a local service, add authentication to an endpoint, receive webhooks, and troubleshoot ngrok errors.

​
Give your AI agent the ngrok documentation

Your AI agent can read the ngrok documentation in a format built for language models:
  • llms.txt: https://ngrok.com/docs/llms.txt lists every docs page with a short description.
  • Full text: https://ngrok.com/docs/llms-full.txt contains the full text of the documentation in one file.
  • Markdown pages: Add .md to the URL of any docs page to get it as Markdown, for example https://ngrok.com/docs/start.md.
  • Docs MCP server: Add https://ngrok.com/docs/mcp to your AI agent as a remote MCP server to let it search the documentation. This server only searches public docs and can’t access your account.