lahfir/agent-desktop
Native desktop automation CLI for AI agents. Control any application through OS accessibility trees with structured JSON output and deterministic element refs.
This tool helps automate tasks across any native desktop application on macOS, allowing you to interact with software like Safari, Slack, or System Settings programmatically. It takes commands to observe, click, type, or manage windows, and outputs structured JSON responses, enabling you to build AI agents that can control your computer. Developers and AI engineers can use this to create automated workflows and intelligent assistants that interact with desktop applications.
Use this if you need to build AI agents or automated scripts that interact directly with macOS desktop applications through their accessibility features, rather than just web browsers.
Not ideal if you are looking for a visual, low-code automation builder or if you primarily need to automate tasks within web browsers or specific, well-supported applications with existing APIs.
Stars
32
Forks
3
Language
Rust
License
—
Category
Last pushed
Mar 11, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/lahfir/agent-desktop"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
open-webui/open-terminal
A computer you can curl ⚡
pproenca/agent-tui
TUI automation for AI agents. Control any terminal app from code.
AI-Shell-Team/aish
Empower the Shell to think. Evolve Operations.
VritraSecz/VritraAI
🚀 Next-gen AI terminal shell: command explanations, code analysis, security scanning, 37 themes,...
makefinks/daemon
Terminal AI agent with a sci-fi avatar: OpenTUI interface, text/voice, hotkey-driven