Download
AI-Hands is a free, open-source Windows automation server that enables AI agents to interact with desktop applications and web browsers. Built with Rust, it exposes automation capabilities through the Model Context Protocol. Its browser layer uses the Chrome DevTools Protocol to navigate websites and interact with page elements. Windows UI Automation provides structured access to application controls and accessibility information. A visual fallback combines screenshots, OCR, and template matching for interfaces without accessible controls. The system supports clicking, typing, window inspection, and other desktop interactions. Monitor restrictions and configurable tool profiles help control automation permissions. Compatible clients include Claude Desktop, Claude Code, Codex CLI, and Gemini CLI.
Features
- Multi-layer AI desktop automation combining browser, Windows UIA, and visual controls
- Chrome DevTools Protocol integration for structured browser navigation and interaction
- Native Windows accessibility tree inspection with element-based clicking and typing
- Screenshot capture, optical character recognition, and visual template matching
- MCP server compatibility with Claude, Codex, Gemini, and other AI agent clients
- Configurable tool permissions, monitor restrictions, and optional voice integration
Project Samples
