[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72451":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":36,"readmeContent":37,"aiSummary":38,"trendingCount":16,"starSnapshotCount":16,"syncStatus":39,"lastSyncTime":40,"discoverSource":41},72451,"mobile-use","minitap-ai\u002Fmobile-use","minitap-ai","AI agents can now use real Android and iOS apps, just like a human.","https:\u002F\u002Fminitap.ai",null,"Python",2586,221,15,9,0,6,17,62,18,29.04,"Apache License 2.0",false,"main",true,[27,28,29,30,31,32,33,5,34,35],"agents","ai","browser-use","langchain","langgraph","langgraph-python","mobile","python","qa","2026-06-12 02:03:03","# mobile-use: automate your phone with natural language\n\n\u003Cdiv align=\"center\">\n\n![mobile-use in Action](.\u002Fdoc\u002Fbanner-v2.png)\n\n\u003C\u002Fdiv>\n\n\u003Cdiv align=\"center\">\n\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fdiscord\u002F1403058278342201394?color=7289DA&label=Discord&logo=discord&logoColor=white&style=for-the-badge)](https:\u002F\u002Fdiscord.gg\u002F6nSqmQ9pQs)\n[![GitHub stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fminitap-ai\u002Fmobile-use?style=for-the-badge&color=e0a8dd)](https:\u002F\u002Fgithub.com\u002Fminitap-ai\u002Fmobile-use\u002Fstargazers)\n\n\u003Ch3>\n    \u003Ca href=\"https:\u002F\u002Fplatform.mobile-use.ai\">\u003Cb>☁️ Cloud\u003C\u002Fb>\u003C\u002Fa> •\n    \u003Ca href=\"https:\u002F\u002Fdocs.minitap.ai\u002Fv2\u002Fmcp-server\u002Fintroduction\">\u003Cb>📚 Documentation\u003C\u002Fb>\u003C\u002Fa> •\n    \u003Ca href=\"https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.07787\">\u003Cb>📃 Paper\u003C\u002Fb>\u003C\u002Fa>\n\n\u003C\u002Fh3>\n\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002F6nSqmQ9pQs\">\u003Cb>Discord\u003C\u002Fb>\u003C\u002Fa> •\n    \u003Ca href=\"https:\u002F\u002Fx.com\u002Fminitap_ai?t=iRWtI497UhRGLeCKYQekig&s=09\">\u003Cb>Twitter \u002F X\u003C\u002Fb>\u003C\u002Fa>\n\u003C\u002Fp>\n\n[![PyPI version](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fminitap-mobile-use.svg?color=blue)](https:\u002F\u002Fpypi.org\u002Fproject\u002Fminitap-mobile-use\u002F)\n[![Python Version](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fpython-3.12%2B-blue)](https:\u002F\u002Fwww.python.org\u002Fdownloads\u002F)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-Apache%202.0-blue)](https:\u002F\u002Fgithub.com\u002Fminitap-ai\u002Fmobile-use\u002Fblob\u002Fmain\u002FLICENSE)\n\n\u003C\u002Fdiv>\n\nMobile-use is a powerful, open-source AI agent that controls your Android or IOS device using natural language. It understands your commands and interacts with the UI to perform tasks, from sending messages to navigating complex apps.\n\n> Mobile-use is quickly evolving. Your suggestions, ideas, and reported bugs will shape this project. Do not hesitate to join in the conversation on [Discord](https:\u002F\u002Fdiscord.gg\u002F6nSqmQ9pQs) or contribute directly, we will reply to everyone! ❤️\n\n## ✨ Features\n\n- 🗣️ **Natural Language Control**: Interact with your phone using your native language.\n- 📱 **UI-Aware Automation**: Intelligently navigates through app interfaces (note: currently has limited effectiveness with games as they don't provide accessibility tree data).\n- 📊 **Data Scraping**: Extract information from any app and structure it into your desired format (e.g., JSON) using a natural language description.\n- 🔧 **Extensible & Customizable**: Easily configure different LLMs to power the agents that power mobile-use.\n\n## Benchmarks\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fminitap.ai\u002Fbenchmark\">\n    \u003Cimg src=\"https:\u002F\u002Ffiles.peachworlds.com\u002Fwebsite\u002F2b590171-669d-42ce-b4b5-ce6eae83a9d8\u002Fscorerank-140126.webp\" alt=\"Project banner\" \u002F>\n  \u003C\u002Fa>\n\u003C\u002Fp>\n\nWe stand as the top performers and the first to have completed 100% of the AndroidWorld benchmark.\n\nGet more info about how we reached this milestone here: [Minitap Benchmark](https:\u002F\u002Fminitap.ai\u002Fbenchmark).\n\nThe official leaderboard is available [here](https:\u002F\u002Fdocs.google.com\u002Fspreadsheets\u002Fd\u002F1cchzP9dlTZ3WXQTfYNhh3avxoLipqHN75v1Tb86uhHo\u002Fedit?pli=1&gid=0#gid=0).\n\nCheck out our research paper [here](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.07787).\n\n## 🚀 Getting Started\n\nReady to automate your mobile experience? Follow these steps to get mobile-use up and running.\n\n### 🌐 From our Platform\n\nEasiest way to get started is to use our Platform.\nFollow our [Platform quickstart](https:\u002F\u002Fdocs.minitap.ai\u002Fmobile-use-sdk\u002Fplatform-quickstart) to get started.\n\n### 🛠️ From source\n\n1.  **Set up Environment Variables:**\n    Copy the example `.env.example` file to `.env` and add your API keys.\n\n    ```bash\n    cp .env.example .env\n    ```\n\n2.  **(Optional) Customize LLM Configuration:**\n    To use different models or providers, create your own LLM configuration file.\n\n    ```bash\n    cp llm-config.override.template.jsonc llm-config.override.jsonc\n    ```\n\n    Then, edit `llm-config.override.jsonc` to fit your needs.\n\n    You can also use local LLMs or any other openai-api compatible providers :\n\n    1. Set `OPENAI_BASE_URL` and `OPENAI_API_KEY` in your `.env`\n    2. In your `llm-config.override.jsonc`, set `openai` as the provider for the agent nodes you want, and choose a model supported by your provider.\n\n    > [!NOTE]\n    > If you want to use Google Vertex AI, you must either:\n    >\n    > - Have credentials configured for your environment (gcloud, workload identity, etc…)\n    > - Store the path to a service account JSON file as the GOOGLE_APPLICATION_CREDENTIALS environment variable\n    >\n    > More information: - [Credential types](https:\u002F\u002Fcloud.google.com\u002Fdocs\u002Fauthentication\u002Fapplication-default-credentials#GAC) - [google.auth API reference](https:\u002F\u002Fgoogleapis.dev\u002Fpython\u002Fgoogle-auth\u002Flatest\u002Freference\u002Fgoogle.auth.html#module-google.auth)\n\n### Quick Launch (Docker)\n\n> [!NOTE]\n> This quickstart, is only available for Android devices\u002Femulators as of now, and you must have Docker installed.\n\nFirst:\n\n- Either plug your Android device and enable USB-debugging via the Developer Options\n- Or launch an Android emulator\n\nThen run in your terminal:\n\n1. For Linux\u002FmacOS:\n\n```bash\nchmod +x mobile-use.sh\nbash .\u002Fmobile-use.sh \\\n  \"Open Gmail, find first 3 unread emails, and list their sender and subject line\" \\\n  --output-description \"A JSON list of objects, each with 'sender' and 'subject' keys\"\n```\n\n2. For Windows (inside a Powershell terminal):\n\n```powershell\npowershell.exe -ExecutionPolicy Bypass -File mobile-use.ps1 `\n  \"Open Gmail, find first 3 unread emails, and list their sender and subject line\" `\n  --output-description \"A JSON list of objects, each with 'sender' and 'subject' keys\"\n```\n\n> [!NOTE]\n> If using your own device, make sure to accept the ADB-related connection requests that will pop up on your device.\n\n#### 🧰 Troubleshooting\n\nThe script will try to connect to your device via IP.\nTherefore, your device **must be connected to the same Wi-Fi network as your computer**.\n\n##### 1. No device IP found\n\nIf the script fails with the following message:\n\n```\nCould not get device IP. Is a device connected via USB and on the same Wi-Fi network?\n```\n\nThen it couldn't find one of the common Wi-Fi interfaces on your device.\nTherefore, you must determine what WLAN interface your phone is using via `adb shell ip addr show up`.\nThen add the `--interface \u003CYOUR_INTERFACE_NAME>` option to the script.\n\n##### 2. Failed to connect to \u003CDEVICE_IP>:5555 inside Docker\n\nThis is most probably an issue with your firewall blocking the connection. Therefore there is no clear fix for this.\n\n##### 3. Failed to pull GHCR docker images (unauthorized)\n\nSince UV docker images rely on a `ghcr.io` public repositories, you may have an expired token if you used `ghcr.io` before for private repositories.\nTry running `docker logout ghcr.io` and then run the script again.\n\n### Manual Launch (Development Mode)\n\nFor developers who want to set up the environment manually:\n\n#### 1. Device Support\n\nMobile-use currently supports the following devices:\n\n- **Physical Android Phones**: Connect via USB with USB debugging enabled.\n- **Android Simulators**: Set up through Android Studio.\n- **iOS Simulators**: Supported for macOS users.\n\n> [!NOTE]\n> Physical iOS devices are not yet supported.\n\n#### 2. Prerequisites\n\n**For Android:**\n\n- **[Android Debug Bridge (ADB)](https:\u002F\u002Fdeveloper.android.com\u002Fstudio\u002Freleases\u002Fplatform-tools)**: A tool to connect to your device.\n\n**For iOS (macOS only):**\n\n- **[Xcode](https:\u002F\u002Fdeveloper.apple.com\u002Fxcode\u002F)**: Apple's IDE for iOS development.\n- **[fb-idb](https:\u002F\u002Ffbidb.io\u002Fdocs\u002Finstallation\u002F)**: Facebook's iOS Development Bridge for device automation.\n\n  ```bash\n  # Install via Homebrew (macOS)\n  brew tap facebook\u002Ffb\n  brew install idb-companion\n  ```\n\n  > [!NOTE]\n  > `idb_companion` is required to communicate with iOS simulators. Make sure it's in your PATH after installation.\n\n**Common requirements:**\n\nBefore you begin, ensure you have the following installed:\n\n- **[uv](https:\u002F\u002Fgithub.com\u002Fastral-sh\u002Fuv)**: A lightning-fast Python package manager.\n\n#### 3. Installation\n\n1.  **Clone the repository:**\n\n    ```bash\n    git clone https:\u002F\u002Fgithub.com\u002Fminitap-ai\u002Fmobile-use.git && cd mobile-use\n    ```\n\n2.  [**Setup environment variables**](#-getting-started)\n\n3.  **Create & activate the virtual environment:**\n\n    ```bash\n    # This will create a .venv directory using the Python version in .python-version\n    uv venv\n\n    # Activate the environment\n    # On macOS\u002FLinux:\n    source .venv\u002Fbin\u002Factivate\n    # On Windows:\n    .venv\\Scripts\\activate\n    ```\n\n4.  **Install dependencies:**\n    ```bash\n    # Sync with the locked dependencies for a consistent setup\n    uv sync\n    ```\n\n## 👨‍💻 Usage\n\nTo run mobile-use, simply pass your command as an argument.\n\n**Example 1: Basic Command**\n\n```bash\npython .\u002Fsrc\u002Fmobile_use\u002Fmain.py \"Go to settings and tell me my current battery level\"\n```\n\n**Example 2: Data Scraping**\n\nExtract specific information and get it back in a structured format. For instance, to get a list of your unread emails:\n\n```bash\npython .\u002Fsrc\u002Fmobile_use\u002Fmain.py \\\n  \"Open Gmail, find all unread emails, and list their sender and subject line\" \\\n  --output-description \"A JSON list of objects, each with 'sender' and 'subject' keys\"\n```\n\n> [!NOTE]\n> If you haven't configured a specific model, mobile-use will prompt you to choose one from the available options.\n\n## 🔎 Agentic System Overview\n\n\u003Cdiv align=\"center\">\n\n![Graph Visualization](doc\u002Fgraph.png)\n\n_This diagram is automatically updated from the codebase. This is our current agentic system architecture._\n\n\u003C\u002Fdiv>\n\n## ❤️ Contributing\n\nWe love contributions! Whether you're fixing a bug, adding a feature, or improving documentation, your help is welcome. Please read our **[Contributing Guidelines](CONTRIBUTING.md)** to get started.\n\n## ⭐ Star History\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fstar-history.com\u002F#minitap-ai\u002Fmobile-use&Date\">\n    \u003Cimg src=\"https:\u002F\u002Fapi.star-history.com\u002Fsvg?repos=minitap-ai\u002Fmobile-use&type=Date\" alt=\"Star History Chart\" \u002F>\n  \u003C\u002Fa>\n\u003C\u002Fp>\n\n## 🏆 Attribution & Licensing\n\n`mobile-use` is the first agentic framework to achieve **100% on the AndroidWorld benchmark**.\n\nThis project is licensed under the **Apache License 2.0**.\n\nIf you use this code, or are inspired by the architecture used to reach our benchmark results, we kindly request that you credit Minitap, Inc.\n\n### How to Cite\nIf you use this work in research or a commercial product, please use the following:\n> Pierre-Louis Favreau, Jean-Pierre Lo, Clement Guiguet, Charles Simon-Meunier,  \nNicolas Dehandschoewercker, Allen G. Roush, Judah Goldfeder, Ravid Shwartz-Ziv.  \n_Do Multi-Agents Dream of Electric Screens? Achieving Perfect Accuracy on AndroidWorld Through Task Decomposition._  \narXiv preprint arXiv:2602.07787 (2026).  \nhttps:\u002F\u002Farxiv.org\u002Fabs\u002F2602.07787\n\n#### Bibtex\n\n\n```bibtex\n@misc{favreau2026multiagentsdreamelectricscreens,\n  title        = {Do Multi-Agents Dream of Electric Screens? Achieving Perfect Accuracy on AndroidWorld Through Task Decomposition},\n  author       = {Pierre-Louis Favreau and Jean-Pierre Lo and Clement Guiguet and Charles Simon-Meunier and Nicolas Dehandschoewercker and Allen G. Roush and Judah Goldfeder and Ravid Shwartz-Ziv},\n  year         = {2026},\n  eprint       = {2602.07787},\n  archivePrefix= {arXiv},\n  primaryClass = {cs.AI},\n  url          = {https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.07787}\n}\n\n","mobile-use 是一个开源的AI代理项目，能够让AI像人一样使用真实的Android和iOS应用。其核心功能包括通过自然语言控制手机、智能地与应用程序界面交互以及从应用中抓取数据并按需格式化。该项目支持多种大型语言模型（LLM），便于用户根据需求自定义配置。mobile-use 适用于需要自动化日常手机操作、提高工作效率或进行移动应用测试等场景。它基于Python开发，并遵循Apache License 2.0开源许可协议，鼓励社区成员参与贡献。",2,"2026-06-11 03:42:06","high_star"]