A production-ready, autonomous AI agent for Android that executes complex tasks through self-planning and screen automation. It features a vision engine for real-time screen analysis, a four-tier memory system, and supports local or remote LLMs with automatic failover.
- Built with
- Gemini Omni FlashNEWSonnet 5NEWSeed 2.1 ProNEWStrong
The source explicitly names a specific model that was verified and added to the registry.
source explicitly credits Gemini 2.0 Flash (README lists 'Gemini 2.0 Flash, Pro, Nano' as supported providers); Claude Sonnet 4 (README lists 'Claude Sonnet 4, Opus 4' as supported providers); DeepSeek V3 (README lists 'DeepSeek V3, R1' as supported providers); GPT-4.1 (README lists 'GPT-4o, GPT-4.1, o3' as supported providers)
- Posted by
- Yashab Alam@yashab-cyberWe haven't verified this person made this build — attribution is our best guess from public sources.
- Discovered
- Discovered onX
- Source date
- Published on X Jul 5, 2026
- Listed
- Added to Dropday 4h ago
- Evidence
- Strong
Comprehensive GitHub repository with 88 commits, detailed architecture diagrams, screenshots of a working UI, and a provided APK distribution link.