AI Agent · Open Source · Android · Decrypt
This Open-Source Phone AI Agent Sees, Hears and Acts—All Without Touching the Cloud
Compiled by KHAO Editorial — aggregated from 1 source. See llms.txt for citation guidance.
★ Tier-1 Source
Your phone already has a camera, a microphone, and a screen.
Key facts
- X-OmniClaw's overall architecture is based on three pillars: Omni Perception, Omni Action, and Omni Memory that work as one continuous loop, with cloud LLMs called in only for heavy reasoning
- Hermes Agent by Nous Research took things further with a self-improving learning loop that compounds capabilities over time
- Omni Memory is what separates X-OmniClaw from a one-shot chatbot
- They run on cloud servers that host virtual copies of Android, letting an AI tap and scroll through apps remotely
Summary
The framework builds a long-term semantic memory from your photo gallery and session history, letting it act as a continuous assistant rather than a one-shot chatbot. A behavior cloning feature lets users record a navigation path once so the agent can replay it instantly via Android deeplink, bypassing multi-step app navigation in future sessions. That project is X-OmniClaw, published by the Multi-X Team. Most mobile AI systems don't run on your phone.