Demo: MobileClaw: A Notification-Driven, Human-Gated Mobile LLM Agent
Abstract
Smartphone assistants built on large language models (LLMs) today either drive an application’s screen display or wait to be asked through chatting; neither is integrated into the event stream produced by the mobile OS and user-installed applications. We demonstrate MobileClaw, an Android system component that attaches an external LLM to the OS notification channel. A background service captures every posted notification and asks the model for recommended further actions; an interactive front end carries out the user’s requests through built-in skills, operated through OS API calls rather than screen automation. The model is treated as an untrusted component: nothing it decides takes effect unless the user confirms it. Notification text and everything the app reads from the device pass a sanitizer before they leave the phone, and privileged operations run in a separate process behind an allowlist it re-checks itself. Demo attendees may fire synthetic or real notifications, manipulate local resources on the device, approve or reject suggested actions, and watch the model’s reasoning trace in real time.
An agent workspace on your phone
MobileClaw is an open-source Android research prototype that connects an LLM to the phone’s notification channel. Incoming notifications can initiate assistance, and interactive chat uses the same conversations, memory and local resource lookup. The phone coordinates the agent loop and executes supported operations through notification actions, Android APIs and AppFunctions exposed by cooperating apps, with user confirmation by default for outward actions.
The model service is configurable: use a cloud API or an OpenAI-compatible local server, including one running on the phone. For example, a photo request can trigger a local search and present candidate images. The user selects a photo, requests sharing, and confirms the image and recipient.
Watch the demo
Download the video (MP4, 1920 × 1080, 2 min 59 sec; English narration and subtitles). Download the English subtitles (SRT). The GitHub repository includes source code, installable APKs, documentation and instructions for trying the demo.
Poster
Alexander Du, Ozzie Hassel and Jash Uganandan contributed equally to this work.
