AI Agents
page-agent
Alibaba's in-page GUI agent: control any web interface with natural language, from one script tag
★ 29.4k starsMIT
View guide →Control Android and iOS devices with natural language — an LLM-agnostic mobile agent framework and CLI.
mobilerun is an open-source ai agents project by droidrun (droidrun/mobilerun on GitHub). It has 9,590 GitHub stars and 1,021 forks, is released under the MIT license, and was last updated on GitHub on 5 October 2026.
Decide the task in plain language, which device (physical, emulator or cloud) you will use, and which LLM provider will drive the agent.
Clone droidrun/mobilerun from GitHub (or use the install path in the run steps below) and skim its README for the parts relevant to your goal.
Install the mobilerun package, install the Portal app on the device, and configure your LLM provider credentials.
Verify the device connection, then run a single natural-language command — for example opening an app and completing a simple flow — and inspect the trace.
Run one-off tasks from the CLI in plain language, or build custom workflows with the Python API and custom tools.
Control Android through the Portal app, or target iOS through the iOS Portal flow.
The agent uses UI trees, screenshots, text input, gestures, app launching and device state from the Portal runtime.
--vision sends screenshots to the LLM, --vision-only controls by screenshot alone, and --reasoning enables manager-executor planning for longer tasks.
Debug executions with Arize Phoenix, Langfuse, saved trajectories and detailed logs.
Return structured data from mobile workflows, and add app-specific guidance cards so the agent performs better on your apps.
Get the source from the official GitHub repository for mobilerun.
git clone https://github.com/droidrun/mobilerun.git
cd mobilerunInstall the package with uv as the README's installation section shows, then install the Portal on your device.
uv pip install mobilerunAfter verifying the Portal connection and configuring your LLM provider, run a natural-language task from the CLI.
mobilerun "Open Settings and turn on Wi-Fi" # see README quickstart for exact CLI syntaxThe droidrun team, through the droidrun/mobilerun repository, with the framework site at mobilerun.ai.
Yes — the project is open source under the MIT license as recorded for this listing. You can use, modify and distribute it under the MIT terms.
The framework is LLM-agnostic but needs an LLM you configure — typically your own API key for a provider, billed by that provider; Mobilerun itself is free.
Appium scripts specify every tap and selector and break when UIs change. Mobilerun describes the goal in natural language and lets an LLM plan through the live UI tree and screenshots — more flexible, but dependent on model quality and cost per run.
Clone the repository (or use the documented installer), follow the install and configure steps on this page, and run the smallest real task first — verify the device connection, then run a single natural-language command — for example opening an app and completing a simple flow — and inspect the trace.
Alibaba's in-page GUI agent: control any web interface with natural language, from one script tag
Android GUI agent framework: AI that sees, plans, and taps real mobile apps.
🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming
Multi-agent LLM framework that mirrors a trading firm: analysts, researchers, trader and risk team
AI agents that use real Android and iOS apps like a human — from minitap.
Microsoft's GUI agent that grew into UFO³ — orchestrating agents across Windows, Linux and mobile devices.
The most-starred other AI Agents projects in the LibreAI catalogue — the realistic shortlist if mobilerun is not the right fit.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Free and open — grab the source, star the repo, and start creating.