# Browserbase (MCP Server)

> Cloud browser automation for agents powered by Stagehand

- Type: MCP server
- Publisher: Browserbase (official)
- Categories: Development, Productivity
- Website: https://docs.browserbase.com/integrations/mcp/setup
- Docs: https://docs.browserbase.com/integrations/mcp/introduction
- Source: https://github.com/browserbase/mcp-server-browserbase
- Transport: streamable-http
- Server URL: https://mcp.browserbase.com/mcp
- Launch command: npx @browserbasehq/mcp
- Sign-in: api-key
- Environment variables: BROWSERBASE_API_KEY, BROWSERBASE_PROJECT_ID
- GitHub stars: 3411
- Listed: 2024-12-05

## About

Browserbase's MCP server gives agents a cloud browser controlled with natural language via Stagehand. Agents can start sessions, navigate, act, observe elements and extract data. The hosted server is recommended; the original self-hosted repository was archived in 2026.

## Install

1. **Connect it in ChatGPT** Turn on Developer mode in ChatGPT (Settings → Security and login), open Plugins, select + and paste https://mcp.browserbase.com/mcp as the MCP server URL. Add it to a chat from the composer's tools menu.

2. **Or add it to Codex or the API**

   Codex CLI:

   ```bash
   codex mcp add browserbase --url https://mcp.browserbase.com/mcp --bearer-token-env-var BROWSERBASE_API_KEY
   ```

   config.toml:

   ```toml
   [mcp_servers.browserbase]
   url = "https://mcp.browserbase.com/mcp"
   bearer_token_env_var = "BROWSERBASE_API_KEY"
   ```

   Responses API:

   ```python
   import os
   from openai import OpenAI
   
   client = OpenAI()
   
   response = client.responses.create(
       model="gpt-6-astra",
       tools=[{
           "type": "mcp",
           "server_label": "browserbase",
           "server_url": "https://mcp.browserbase.com/mcp",
           "authorization": os.environ["BROWSERBASE_API_KEY"],
           "require_approval": "never",
       }],
       input="Go to news.ycombinator.com and extract the top 10 story titles",
   )
   print(response.output_text)
   ```

## Tools

- `start`
- `navigate`
- `act`
- `observe`
- `extract`
- `end`

## Example prompts

- "Go to news.ycombinator.com and extract the top 10 story titles"
- "Fill out the contact form on this page with test data"
- "Log into the staging site and screenshot the dashboard"

## Related

- [Chrome DevTools MCP](https://chatgptplugins.directory/mcp/chrome-devtools.md): Let coding agents control and debug a live Chrome browser
- [Playwright MCP](https://chatgptplugins.directory/mcp/playwright.md): Browser automation for agents via Playwright accessibility snapshots
- [GitHub](https://chatgptplugins.directory/mcp/github.md): Manage repos, issues, pull requests and Actions from your AI agent
- [Context7](https://chatgptplugins.directory/mcp/context7.md): Pull up-to-date, version-specific library docs into your coding agent
- [Notion](https://chatgptplugins.directory/mcp/notion.md): Search, read, and write Notion pages, databases, and comments via hosted MCP
- [Excalidraw](https://chatgptplugins.directory/mcp/excalidraw.md): Draw hand-drawn Excalidraw diagrams live inside ChatGPT, Claude and VS Code

---
Source: https://chatgptplugins.directory/mcp/browserbase · Index: https://chatgptplugins.directory/llms.txt · Independent and community-run. Not affiliated with or endorsed by OpenAI. ChatGPT is a trademark of OpenAI.