Browser Use offers a hosted MCP server that starts and manages cloud browser-agent sessions authenticated with an API key header, and an open-source local stdio server that exposes direct browser controls (navigate, click, type, extract, tabs) plus an autonomous agent tool. The local server runs via uvx and needs an OpenAI or Anthropic key for agent tasks.
Also available as
Connect
- 1Connect it in ChatGPT
Turn on Developer mode in ChatGPT (Settings → Security and login), open Plugins, select + and paste https://api.browser-use.com/v3/mcp as the MCP server URL. Add it to a chat from the composer's tools menu.
- 2Or add it to Codex or the API
codex mcp add browser-use --url https://api.browser-use.com/v3/mcp --bearer-token-env-var OPENAI_API_KEY
Configs are generated from the publisher's documented endpoint. Check their docs for plan requirements and scopes. New to this? Read the step-by-step tutorial.
Tools 10
run_sessionget_sessionsend_taskstop_sessionlist_sessionsbrowser_navigatebrowser_clickbrowser_typebrowser_extract_contentretry_with_browser_use_agentThings to ask
- “Go to the pricing page of three competitors and compare their plans”
- “Fill out the contact form on example.com with test data”
- “Check whether the signup flow on staging works end to end”
- “Extract all job listings from this careers page”