AOHP: An Open-Source OS-Level Agent Harness for Personalized, Efficient and Secure Interaction

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Shanhui Zhao, Jiacheng Liu, Guohong Liu, Jichao Yan, Jialei Ye, Yuhao Yang, Hao Wen, Shizuo Tian, Yizhen Yuan, Yuxuan Chen, Yunxin Liu, Ju Ren, Ya-Qin Zhang, Chao Huang, Yao Guo, Yuanchun Li
• Venue: arXiv: cs.AI
• Published: 2026-06-22
• Collected: 2026-06-23
• URL: http://arxiv.org/abs/2606.23449v1
• PDF: https://arxiv.org/pdf/2606.23449v1
• DOI: UNKNOWN

Plain-Language Brief

The paper presents the Android Open Harness Project (AOHP), an operating system level agent harness built on the Android Open Source Project. AOHP aims to treat AI agents as first-class OS actors, enabling adaptive user interfaces and agent-friendly runtime environments. It introduces three mechanisms: personalized service composition, efficient agent interfaces, and secure information flow. Preliminary experiments suggest that AOHP improves task completion rates, reduces execution cost, and enhances security-policy compliance.

What the Document Says

• The paper presents a new operating system level agent harness called AOHP built on the Android Open Source Project (AOSP).
• AOHP aims to enable adaptive user interfaces and agent-friendly runtime environments by treating AI agents as first-class OS actors.
• Three mechanisms have been introduced in AOHP: personalized service composition, efficient agent interfaces, and secure information flow.
• The experiments conducted on challenging tasks show that AOHP improves task completion rates by 21.12%, reduces execution cost by 51.55%, and enhances security-policy compliance.

Key Concepts Mentioned

• AI agents
• operating system
• Android Open Source Project (AOSP)
• adaptive user interfaces
• agent-friendly runtime environments
• personalized service composition
• efficient agent interfaces
• secure information flow
• task completion rate
• execution cost
• security-policy compliance

Claims Made by the Document

• AOHP improves task completion rates, reduces execution cost, and enhances security-policy compliance.

Evidence Used

• Based on preliminary experiments on challenging tasks, AOHP shows clear advantages in task completion (+21.12% completion rate), execution cost (-51.55% token cost), and security-policy compliance.

Methods / Approach

• Preliminary experiments were conducted on challenging tasks covering key capabilities of OS agents.

Limitations Stated by the Authors

The paper does not provide details about the specific nature of the challenges or the experimental setup.

Limitations Noted by the Agent

Agent interpretation — not from the source.

The agent is based on the Android operating system, which may limit its applicability to other platforms.

Open Questions

• What are the specific challenges and experimental setup used in the preliminary experiments?
• How does AOHP perform on other operating systems?

Notes for Future Q&A

May include interpretation. Labeled where it does.

The paper provides no details about the specific nature of the challenges or the experimental setup. The agent is based on the Android operating system, which may limit its applicability to other platforms.

Raw Abstract / Excerpt

AI agents are driving a new software paradigm, with the ability to autonomously call tools, extract information, manage memory, and complete tasks that span applications and data sources. Most existing end-user operating systems, however, are designed for application-centric workflows and offer little native support for AI agents. This mismatch limits the wider adoption of agents and leads to execution overhead and safety risks when running agents on conventional systems. While the concept of agent-native operating systems is emerging, the research community lacks an open testbed to explore the architectural primitives desired for agent-mediated interaction. We present AOHP (Android Open Harness Project), an OS-level agent harness built on the Android Open Source Project (AOSP). The core design principle of AOHP is to treat agents as first-class OS actors, enabling adaptive user interfaces and agent-friendly runtime environments. AOHP preserves the mature Android software and hardware ecosystem while introducing three agent-oriented system mechanisms: personalized service composition, efficient agent interfaces, and secure information flow. Based on preliminary experiments on challenging tasks covering key capabilities of OS agents, AOHP shows clear advantages in task completion (+21.12% completion rate), execution cost (-51.55% token cost), and security-policy compliance.