Tag
MobileRun claims a perfect 116/116 score on Google's AndroidWorld benchmark, covering 20 Android apps and multi-step workflows, reportedly outperforming Google's Artemis agent at 99%+.
The article accuses Google's Artemis project of copying code from Minitap's open-source mobile-use without attribution, removing original author names, and raising concerns about open-source ethics.
The author announces mobile-harness, an open-source agent harness that provides a unified API for AI agents to control real iOS and Android devices, allowing any agent to automate mobile interactions.
This article explains how to connect OpenClaw, an AI orchestrator, to a real phone using the MCP protocol and Mobilerun, an open-source tool for real-device automation, enabling tasks that require device integrity or mobile-only apps.
Explains an AI agent phone as a real or cloud-hosted phone operated by an LLM agent that interacts with mobile-only apps via accessibility tree and vision, highlighting limitations and calling for community input on reliability.
The author argues that hardware-level interaction (HDMI capture + USB HID touch simulation) is superior to ADB-based automation for building OS-agnostic general-purpose AI agents, and introduces their project Aiden using a cheap RV1106 dev board.
QApilot's CoWork claims to triple mobile automation efficiency without expanding the QA team.
H Company releases Holo3.1, a family of Vision-Language Models (0.8B to 35B) for computer use agents, supporting web, desktop, and mobile automation with native function calling and optimized quantized checkpoints for local deployment.
Holo3.1 is an updated computer-use model family that improves robustness across web, desktop, and mobile environments, introduces quantized checkpoints for local execution, and adds native support for function-calling protocols.
OpenGUI is highlighted as a novel AI agent platform that utilizes actual Android devices for task execution, offering a more realistic interface than traditional browser-based agents.
William is introduced as the first AI agent that operates directly on a phone to automate repetitive tapping, testing and app workflows.
An MCP server for scalable mobile automation and development, enabling AI agents to interact with iOS and Android devices through a platform-agnostic interface.