Turbo1123/roubao

Android Automation Tool Based on Vision-Language Models

49
/ 100
Emerging

Based on the README, here's a technical summary that goes deeper than the one-line description: --- Executes automation tasks entirely on-device by porting MobileAgent's agentic framework to native Kotlin, leveraging Shizuku for system-level permissions without requiring a computer or USB connection. Implements a dual-layer Tools/Skills architecture inspired by Claude Code—delegating to AI-capable apps via DeepLink when available, otherwise performing GUI automation through vision-language model analysis of screenshots captured locally. Supports multiple VLM backends (Qwen, GPT-4V, Claude) with dynamic model discovery and optional local inference endpoints.

1,895 stars.

No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 9 / 25
Community 20 / 25

How are scores calculated?

Stars

1,895

Forks

204

Language

Kotlin

License

MIT

Last pushed

Jan 08, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/agents/Turbo1123/roubao"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.