skills.rest
lahfirlahfirCommunity·2 Agent Skills Included

agent-desktop

Let AI agents see and control any desktop application

Gives AI agents structured access to desktop apps through native OS accessibility trees, with no screenshots or pixel matching. Replaces brittle image-based automation with fast, token-efficient element snapshots and reliable click, type, and scroll commands. Ships a single Rust binary plus C-ABI bindings so agents on macOS, Windows, and Linux automate real apps in seconds.
1,003|64
↗
npx skills add lahfir/agent-desktop --all -g -y

All Skills in This Repository (2)

Pure Emerald Level Indicators
📦 In Repo
lahfirlahfir

agent-desktop-ffi

Provides C-ABI bindings to access agent-desktop's PlatformAdapter for low-level application interaction.

Community
Advanced
📦 In Repo
lahfirlahfir

agent-desktop

Automate desktop application interactions via the agent-desktop CLI.

Community
Advanced

Frequently Asked Questions

FAQPage Schema
How to install agent-desktop?▼

Run `npx skills add lahfir/agent-desktop --all -g -y` in your terminal to install all skills in this suite globally.

How can an AI agent control desktop apps?▼

agent-desktop reads the native accessibility tree of any app and returns structured JSON with element refs, so your agent can click, type, scroll, and manage windows with simple commands.

Does agent-desktop need screenshots or vision models?▼

No. It uses OS accessibility APIs instead of screenshots or pixel matching, which makes automation faster, cheaper, and far more reliable.

Which platforms does agent-desktop support?▼

macOS is fully supported today, with Windows and Linux adapters built against the same core contracts. The same commands work identically across platforms.

Can I call agent-desktop from Python or Node?▼

Yes. Every release ships a prebuilt C-ABI library you can load from Python, Node, Go, Swift, Ruby, or C++ for direct in-process calls instead of spawning the CLI.

Related Repositories in Software Engineering

View All in Software Engineering→
openclawopenclaw
388.4k

openclaw

Run a personal AI assistant across your devices and chat apps

Official📦 120 Skills
obraobra
280.0k

superpowers

Gives coding agents a disciplined workflow from idea to merged code

Community📦 15 Skills
facebookfacebook
248.4k

react

AI agent skills for building, testing, and porting React core

Official📦 14 Skills