ClawMobile: Rethinking Smartphone-Native Agentic Systems

TL;DR

ClawMobile presents a hierarchical architecture for smartphone-native agentic systems that separates reasoning from control, enhancing stability and reproducibility in mobile environments.

Contribution

It introduces a novel hierarchical design for mobile LLM agents, addressing stability and control challenges unique to smartphones.

Findings

Hierarchical architecture improves execution stability on real devices.

Design principles for efficient and adaptable mobile LLM runtimes.

Open-sourced implementation facilitates future research.

Abstract

Smartphones represent a uniquely challenging environment for agentic systems. Unlike cloud or desktop settings, mobile devices combine constrained execution contexts, fragmented control interfaces, and rapidly changing application states. As large language models (LLMs) evolve from conversational assistants to action-oriented agents, achieving reliable smartphone-native autonomy requires rethinking how reasoning and control are composed. We introduce ClawMobile as a concrete exploration of this design space. ClawMobile adopts a hierarchical architecture that separates high-level language reasoning from structured, deterministic control pathways, improving execution stability and reproducibility on real devices. Using ClawMobile as a case study, we distill the design principles for mobile LLM runtimes and identify key challenges in efficiency, adaptability, and stability. We argue that building robust smartphone-native agentic systems demands principled coordination between probabilistic planning and deterministic system interfaces. The implementation is open-sourced~\footnote{https://github.com/ClawMobile/ClawMobile} to facilitate future exploration.