Provides an MCP server to control Chrome or any Chromium browser using accessibility tree snapshots.
Sponsored
PinchTab offers an MCP server that enables AI agents to interact with Chromium-based browsers like Chrome, Brave, Edge, or Yandex. It achieves this by providing accessibility tree snapshots, allowing precise control and data extraction from web pages. Designed for ease of use, PinchTab is automatically included and configured, streamlining the process of integrating powerful browser automation into AI agent workflows without manual setup.
Key Features
01Control any Chromium-based browser (Chrome, Brave, Edge, Yandex)
02Interact with web pages via accessibility tree snapshots
03Zero-setup integration with automatic PinchTab dependency
04Configurable browser binary, port, tokens, and flags
05Supports both headless and visible browser operation
060 GitHub stars
Use Cases
01Integrating with existing browser sessions for tasks requiring login or persistent context
02Empowering AI agents (e.g., Claude) to automate web tasks
03Debugging browser interactions by running in non-headless mode
