{
  "markdown": "# Browserbase MCP Server\n\n[![smithery badge](https://smithery.ai/badge/@browserbasehq/mcp-browserbase)](https://smithery.ai/server/@browserbasehq/mcp-browserbase)\n\n![cover](assets/cover-mcp.png)\n\n[The Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction) is an open protocol that enables seamless integration between LLM applications and external data sources and tools. Whether you're building an AI-powered IDE, enhancing a chat interface, or creating custom AI workflows, MCP provides a standardized way to connect LLMs with the context they need.\n\nThis server provides cloud browser automation capabilities using [Browserbase](https://www.browserbase.com/) and [Stagehand](https://github.com/browserbase/stagehand). It enables LLMs to interact with web pages, take screenshots, extract information, and perform automated actions with atomic precision.\n\n## Features\n\n| Feature            | Description                                                 |\n| ------------------ | ----------------------------------------------------------- |\n| Browser Automation | Control and orchestrate cloud browsers via Browserbase      |\n| Data Extraction    | Extract structured data from any webpage                    |\n| Web Interaction    | Navigate, click, and fill forms with ease                   |\n| Screenshots        | Capture full-page and element screenshots                   |\n| Model Flexibility  | Supports multiple models (OpenAI, Claude, Gemini, and more) |\n| Vision Support     | Use annotated screenshots for complex DOMs                  |\n| Session Management | Create, manage, and close browser sessions                  |\n| Multi-Session      | Run multiple browser sessions in parallel                   |\n\n## How to Setup\n\n### Quickstarts:\n\n#### Add to Cursor\n\nCopy and Paste this link in your Browser:\n\n```text\ncursor://anysphere.cursor-deeplink/mcp/install?name=browserbase&config=eyJjb21tYW5kIjoibnB4IEBicm93c2VyYmFzZWhxL21jcCIsImVudiI6eyJCUk9XU0VSQkFTRV9BUElfS0VZIjoiIiwiQlJPV1NFUkJBU0VfUFJPSkVDVF9JRCI6IiIsIkdFTUlOSV9BUElfS0VZIjoiIn19\n```\n\nWe currently support 2 transports for our MCP server, STDIO and SHTTP. We recommend you use SHTTP with our remote hosted url to take advantage of the server at full capacity.\n\n## SHTTP:\n\nTo use the Browserbase MCP Server through our remote hosted URL, add the following to your configuration.\n\nGo to [smithery.ai](https://smithery.ai/server/@browserbasehq/mcp-browserbase) and enter your API keys and configuration to get a remote hosted URL.\nWhen using our remote hosted server, we provide the LLM costs for Gemini, the [best performing model](https://www.stagehand.dev/evals) in [Stagehand](https://www.stagehand.dev).\n\n![Smithery Image](assets/smithery.jpg)\n\nIf your client supports SHTTP:\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"url\": \"your-smithery-url.com\"\n    }\n  }\n}\n```\n\nIf your client doesn't support SHTTP:\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"command\": \"npx\",\n      \"args\": [\"mcp-remote\", \"your-smithery-url.com\"]\n    }\n  }\n}\n```\n\n## STDIO:\n\nYou can either use our Server hosted on NPM or run it completely locally by cloning this repo.\n\n> **❗️ Important:** If you want to use a different model you have to add --modelName to the args and provide that respective key as an arg. More info below.\n\n### To run on NPM (Recommended)\n\nGo into your MCP Config JSON and add the Browserbase Server:\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"command\": \"npx\",\n      \"args\": [\"@browserbasehq/mcp-server-browserbase\"],\n      \"env\": {\n        \"BROWSERBASE_API_KEY\": \"\",\n        \"BROWSERBASE_PROJECT_ID\": \"\",\n        \"GEMINI_API_KEY\": \"\"\n      }\n    }\n  }\n}\n```\n\nThat's it! Reload your MCP client and Claude will be able to use Browserbase.\n\n### To run 100% local:\n\n```bash\n# Clone the Repo\ngit clone https://github.com/browserbase/mcp-server-browserbase.git\ncd mcp-server-browserbase\n\n# Install the dependencies and build the project\nnpm install && npm run build\n```\n\nThen in your MCP Config JSON run the server. To run locally we can use STDIO or self-host SHTTP.\n\n### STDIO:\n\nTo your MCP Config JSON file add the following:\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"command\": \"node\",\n      \"args\": [\"/path/to/mcp-server-browserbase/cli.js\"],\n      \"env\": {\n        \"BROWSERBASE_API_KEY\": \"\",\n        \"BROWSERBASE_PROJECT_ID\": \"\",\n        \"GEMINI_API_KEY\": \"\"\n      }\n    }\n  }\n}\n```\n\nThen reload your MCP client and you should be good to go!\n\n## Configuration\n\nThe Browserbase MCP server accepts the following command-line flags:\n\n| Flag                       | Description                                                                 |\n| -------------------------- | --------------------------------------------------------------------------- |\n| `--proxies`                | Enable Browserbase proxies for the session                                  |\n| `--advancedStealth`        | Enable Browserbase Advanced Stealth (Only for Scale Plan Users)             |\n| `--keepAlive`              | Enable Browserbase Keep Alive Session                                       |\n| `--contextId <contextId>`  | Specify a Browserbase Context ID to use                                     |\n| `--persist`                | Whether to persist the Browserbase context (default: true)                  |\n| `--port <port>`            | Port to listen on for HTTP/SHTTP transport                                  |\n| `--host <host>`            | Host to bind server to (default: localhost, use 0.0.0.0 for all interfaces) |\n| `--cookies [json]`         | JSON array of cookies to inject into the browser                            |\n| `--browserWidth <width>`   | Browser viewport width (default: 1024)                                      |\n| `--browserHeight <height>` | Browser viewport height (default: 768)                                      |\n| `--modelName <model>`      | The model to use for Stagehand (default: google/gemini-2.0-flash)           |\n| `--modelApiKey <key>`      | API key for the custom model provider (required when using custom models)   |\n| `--experimental`           | Enable experimental features (default: false)                               |\n\nThese flags can be passed directly to the CLI or configured in your MCP configuration file.\n\n### NOTE:\n\nCurrently, these flags can only be used with the local server (npx @browserbasehq/mcp-server-browserbase).\n\n## Configuration Examples\n\n### Proxies\n\nHere are our docs on [Proxies](https://docs.browserbase.com/features/proxies).\n\nTo use proxies, set the --proxies flag in your MCP Config:\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"command\": \"npx\",\n      \"args\": [\"@browserbasehq/mcp-server-browserbase\", \"--proxies\"],\n      \"env\": {\n        \"BROWSERBASE_API_KEY\": \"\",\n        \"BROWSERBASE_PROJECT_ID\": \"\",\n        \"GEMINI_API_KEY\": \"\"\n      }\n    }\n  }\n}\n```\n\n### Advanced Stealth\n\nHere are our docs on [Advanced Stealth](https://docs.browserbase.com/features/stealth-mode#advanced-stealth-mode).\n\nTo use advanced stealth, set the --advancedStealth flag in your MCP Config:\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"command\": \"npx\",\n      \"args\": [\"@browserbasehq/mcp-server-browserbase\", \"--advancedStealth\"],\n      \"env\": {\n        \"BROWSERBASE_API_KEY\": \"\",\n        \"BROWSERBASE_PROJECT_ID\": \"\",\n        \"GEMINI_API_KEY\": \"\"\n      }\n    }\n  }\n}\n```\n\n### Contexts\n\nHere are our docs on [Contexts](https://docs.browserbase.com/features/contexts)\n\nTo use contexts, set the --contextId flag in your MCP Config:\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"@browserbasehq/mcp-server-browserbase\",\n        \"--contextId\",\n        \"<YOUR_CONTEXT_ID>\"\n      ],\n      \"env\": {\n        \"BROWSERBASE_API_KEY\": \"\",\n        \"BROWSERBASE_PROJECT_ID\": \"\",\n        \"GEMINI_API_KEY\": \"\"\n      }\n    }\n  }\n}\n```\n\n### Browser Viewport Sizing\n\nThe default viewport sizing for a browser session is 1024 x 768. You can adjust the Browser viewport sizing with browserWidth and browserHeight flags.\n\nHere's how to use it for custom browser sizing. We recommend to stick with 16:9 aspect ratios (ie: 1920 x 1080, 1280 x 720, 1024 x 768)\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"@browserbasehq/mcp-server-browserbase\",\n        \"--browserHeight 1080\",\n        \"--browserWidth 1920\"\n      ],\n      \"env\": {\n        \"BROWSERBASE_API_KEY\": \"\",\n        \"BROWSERBASE_PROJECT_ID\": \"\",\n        \"GEMINI_API_KEY\": \"\"\n      }\n    }\n  }\n}\n```\n\n### Model Configuration\n\nStagehand defaults to using Google's Gemini 2.0 Flash model, but you can configure it to use other models like GPT-4o, Claude, or other providers.\n\n**Important**: When using any custom model (non-default), you must provide your own API key for that model provider using the `--modelApiKey` flag.\n\nHere's how to configure different models:\n\n```json\n{\n  \"mcpServers\": {\n    \"browserbase\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"@browserbasehq/mcp-server-browserbase\",\n        \"--modelName\",\n        \"anthropic/claude-3-5-sonnet-latest\",\n        \"--modelApiKey\",\n        \"your-anthropic-api-key\"\n      ],\n      \"env\": {\n        \"BROWSERBASE_API_KEY\": \"\",\n        \"BROWSERBASE_PROJECT_ID\": \"\"\n      }\n    }\n  }\n}\n```\n\n_Note: The model must be supported in Stagehand. Check out the docs [here](https://docs.stagehand.dev/examples/custom_llms#supported-llms). When using any custom model, you must provide your own API key for that provider._\n\n### Resources\n\nThe server provides access to screenshot resources:\n\n1. **Screenshots** (`screenshot://<screenshot-name>`)\n   - PNG images of captured screenshots\n\n## Key Features\n\n- **AI-Powered Automation**: Natural language commands for web interactions\n- **Multi-Model Support**: Works with OpenAI, Claude, Gemini, and more\n- **Advanced Session Management**: Single and multi-session support for parallel browser automation\n- **Screenshot Capture**: Full-page and element-specific screenshots\n- **Data Extraction**: Intelligent content extraction from web pages\n- **Proxy Support**: Enterprise-grade proxy capabilities\n- **Stealth Mode**: Advanced anti-detection features\n- **Context Persistence**: Maintain authentication and state across sessions\n- **Parallel Workflows**: Run multiple browser sessions simultaneously for complex automation tasks\n\nFor more information about the Model Context Protocol, visit:\n\n- [MCP Documentation](https://modelcontextprotocol.io/docs)\n- [MCP Specification](https://spec.modelcontextprotocol.io/)\n\nFor the official MCP Docs:\n\n- [Browserbase MCP](https://docs.browserbase.com/integrations/mcp/introduction)\n\n## License\n\nLicensed under the Apache 2.0 License.\n\nCopyright 2025 Browserbase, Inc.\n",
  "bytes": 10775,
  "sha": "f14cb3635d187eb64c797c55fb352f05e38359562593b36124b2ca20a18479ae",
  "repo_slug": "imronai/mcp-server-browserbase",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_ai_smithery_imronai_mcp_server_browserba_5e3480bd/readme"
}