Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

PalmClaw: A Native On-Device Agent Framework for Mobile Phones

Source

arXiv

arxiv.org/abs/2607.13027v1

Read the source

Source

Plain-Language Brief

Researchers have developed a new agent framework called PalmClaw, which allows large language model agents to run natively on mobile phones. This framework aims to overcome the limitations of existing mobile agents that rely on graphical user interface actions and can't directly access device capabilities.

What the Document Says

  • The text discusses a new agent framework named PalmClaw.
  • PalmClaw is designed to run natively on mobile phones.
  • It aims to improve the performance of large language model agents on mobile devices.
  • Existing mobile agents have limitations due to their reliance on graphical user interface actions and inability to directly access device capabilities.
  • PalmClaw addresses these limitations by exposing device capabilities as tools with explicit arguments, structured results, and defined execution boundaries.
  • The design allows agents to use mobile capabilities directly while keeping each action explicit and controlled.
  • Experiments show an improvement in task success and reduction in completion time compared to a strong baseline.
  • The code for PalmClaw is available on GitHub.

Key Concepts Mentioned

  • Large Language Model (LLM) agents
  • PalmClaw agent framework
  • Mobile phones
  • Device capabilities
  • Task automation

Claims Made by the Document

  • PalmClaw improves task success and reduces completion time compared to a strong baseline.
  • PalmClaw addresses limitations of existing mobile agents.

Evidence Used

  • Experiments show an 11.5% relative improvement in task success and a 94.9% reduction in completion time over the strongest baseline.

Methods / Approach

  • Development of PalmClaw agent framework
  • Conducting experiments to compare PalmClaw with a strong baseline

Limitations Stated by the Authors

Not stated in source

Limitations Noted by the Agent

Agent interpretation — not from the source.

The performance improvements and limitations are based on the experiments conducted by the authors.

Open Questions

  • What is the specific strong baseline used for comparison?
  • What other potential limitations or drawbacks of PalmClaw might exist?

Notes for Future Q&A

May include interpretation. Labeled where it does.

It would be useful to know more about the experimental setup, including the specific baseline used and any additional factors that could have influenced the results.

Raw Abstract / Excerpt

Large Language Model (LLM) agents have moved beyond generating responses to executing multi-step tasks by calling tools, observing the results, and iteratively deciding the next action. Most agent systems run on desktops or servers, which support tool use and task automation. Mobile devices are also important agent environments because they are widely accessible and contain users' data, sensors, and daily-use applications. Existing mobile agents mainly operate smartphones through graphical user interface (GUI) actions such as tapping, swiping, and typing, which often form long, interface-dependent sequences, cannot directly access device capabilities, and make execution boundaries difficult to define. We present \textbf{PalmClaw}, an open-source agent framework that runs natively on mobile phones and manages the sessions, memory, skills, tools, and agent loop directly on the device. PalmClaw exposes device capabilities as device tools with explicit arguments, structured results, and clearly defined execution boundaries. This design enables agents to use mobile capabilities directly while keeping each action explicit and controlled. Experiments show an 11.5% relative improvement in task success and a 94.9% reduction in completion time over the strongest baseline, with lower setup burden and traces illustrating how execution boundaries are applied. Code is available at https://github.com/ModalityDance/PalmClaw.