EyeBrowse provides a Python library and MCP server for controlling a stealth Chromium browser with ARIA-ref-based interaction, network inspection, screenshots, PDF/MHTML capture, and other browser automation features. It includes proxy/identity rotation, captcha solver integrations, and can extract rendered pages to markdown without using an LLM.
From the registry: Stealthy, LLM-drivable browser engine — a Python library + 85-tool MCP server on stealth Chromium.

$ pip install eyebrowsehttps://pypi.org/project/eyebrowse/Please install the `eyebrowse` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Python 3.12** — EyeBrowse requires Python 3.12 and is pinned to versions below 3.13. (https://www.python.org/)
- **Install the EyeBrowse package** — Install the MCP server from PyPI. The stealth Chromium browser binary is downloaded automatically on first launch. Run: `pip install eyebrowse` (https://pypi.org/project/eyebrowse/)
Canonical MCP server config (stdio transport):
- command: `eyebrowse-mcp`
- args: []
Note: EyeBrowse ships a FastMCP server over stdio. It runs proxyless by default and the README does not require any environment variables for basic MCP use. Optional proxy or captcha-related settings are mentioned, but no concrete MCP env var names beyond the EYEBROWSE_PROXY_* family are documented in the README.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.EYEBROWSE_PROXY_*Optional default proxy settings for proxy and identity rotationMCP server for Firecrawl web scraping, structured data extraction and web search integration.