#7070th of 118 in Agent frameworks and SDKs
opengui
Android GUI agent framework that operates real phones via screenshots
- Stars
- 1.8k
- License
- custom license
- Last commit
- Nov 2026
- Last release
- Sep 2026
- Language
- HTML
1 on 21 Apr 20261,810 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
OpenGUI lets a vision-capable model see Android screens and perform taps, swipes, typing and navigation on real devices or emulators, for phone tasks and app testing. It ships as a WorkBuddy plugin (macOS), a DeepSeek Harness plugin, or a full self-hosted stack with a backend, Android client, CLI and IM dispatch via Feishu, Telegram and Discord. The source is under BUSL-1.1, so production use needs a commercial license.
Who it is for: Developers automating or testing Android apps with vision-model agents
Strengths
- Plugin install for WorkBuddy or DeepSeek Harness avoids deploying the full backend
- Full stack adds CLI, REST API, and Feishu, Telegram and Discord task dispatch
- Works on real devices over USB debugging, or on a local Android emulator
- Routes to Claude, GPT, Gemini, Kimi, MiniMax or compatible models
Weaknesses
- BUSL-1.1 license: production and commercial use require a paid license; not OSI open source
- WorkBuddy plugin is macOS only and is a public-testing prerelease
- OpenGUI SMS sign-in currently requires a mainland-China phone number
- Android only; needs a model with image input and tool calling
What it needs
- no GPU
- Needs Android device with USB debugging or emulator, ADB / scrcpy, Node.js 22.19+ or 24+ (DSH plugin), WorkBuddy 5.5.3+ or DeepSeek Harness, Redis and PostgreSQL (full stack)
- Models: Claude, GPT, Gemini, Kimi, MiniMax
Quick answers
Is opengui open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Does opengui need a GPU?
No, it runs without one.
Also in Agent frameworks and SDKs
See all 118| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 95 out of 100 |
| 2 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 91 out of 100 |
| 3 | mastraTypeScript framework for building AI agents, workflows and MCP servers | 91 out of 100 |
| #4 | browser-usePython library that lets an LLM agent control a web browser | 90 out of 100 |
| #5 | copilotkitSDK for adding agent chat, generative UI and shared state to apps | 90 out of 100 |