10 open-source GitHub repos that let AI agents control your phone, from Open-AutoGLM to mobile-use
victor_explore · x · 2026-10-18
A curated list of open-source repos for AI agents that operate phones (stars checked Oct 2026):
- Open-AutoGLM (26K★, Apache 2.0): free phone model + framework; type a task, it taps through Android apps; docs mostly in Chinese.
- Maestro (16K★): mobile testing tool with built-in MCP server; your coding agent taps through the app and saves steps as tests.
- Midscene (15K★, MIT): one agent API that sees the screen and acts across Android, iOS, web, desktop.
- UI-TARS (12K★): ByteDance's open screen-seeing model with a phone mode; connect a phone yourself.
- ARTEMIS (11K★, Google): plain-English orders on Android via MCP.
- Mobilerun (10K★), Mobile-Agent (9K★, Alibaba), Mobile MCP (9K★), agent-device (5K★, lets coding agents run real iPhones), mobile-use (3K★, natural-language Android automation).
More from coding & agent
- Coding AI wars: model is faster and smarter but 'dogshit at actual engineering' — MarcJSchmidt · 2026-10-18
- Anthropic borrows lawyers' ethics for Claude agents that closed 186 deals with no human approval — rohanpaul_ai · 2026-10-18
- Prompt injection via email can hijack your AI agent — jev-guard adds risk-scored guards — sujingshen · 2026-10-18
- 14-Agent AI Film Pipeline on One RTX 5090: Fixing Cut Defects Between Clips — developervkmp · 2026-10-18
- Meta paper introduces "agent plasticity": best-performing agent isn't the most efficient learner — rohanpaul_ai · 2026-10-18
- Dev ships a bushfire prediction platform in a weekend with frontier models — CtrlAltDwayne · 2026-10-18