Overview
What is Stagehand MCP Server?
A Model Context Protocol (MCP) server that provides AI-powered web automation capabilities using Stagehand. It enables LLMs to interact with web pages, perform actions, extract data, and observe possible actions in a real browser environment.
How to use Stagehand MCP Server?
Install dependencies (npm install), build the project (npm run build), then configure your Claude Desktop with the required environment variables (BROWSERBASE_API_KEY, BROWSERBASE_PROJECT_ID, OPENAI_API_KEY). After restarting Claude, use the tools (navigate, act, extract, observe, screenshot) via the hammer icon.
Key features of Stagehand MCP Server
- AI-powered web automation in a real browser
- Perform actions on web pages
- Extract structured data from pages
- Observe possible actions on pages
- Simple and extensible API
- Model-agnostic support for various LLM providers
Use cases of Stagehand MCP Server
—
FAQ from Stagehand MCP Server
What environment variables are required?
BROWSERBASE_API_KEY, BROWSERBASE_PROJECT_ID, and OPENAI_API_KEY must be set to authenticate with Browserbase and OpenAI.
What tools does Stagehand MCP Server provide?
The server offers stagehand_navigate, stagehand_act, stagehand_extract, stagehand_observe, and screenshot.
What resources can I access?
Console logs (console://logs) and screenshots (screenshot://<n>) are available as resources.
Is Stagehand MCP Server model-agnostic?
Yes, it supports various LLM providers for flexibility.
How do I enable debug logging?
Set the DEBUG environment variable to enable debug logging.
