🤖 JARVIS

The Ultra-Evolved AI PC Controller

Jarvis is a state-of-the-art Multimodal AI Agent stack that grants you total control over your computer through Natural Language and Voice Commands.

Features • Quick Start • Showcase • Contributing

---

🚀 Unleash the Power of Jarvis

Jarvis transforms your interaction with the digital world. No more complex menus or repetitive tasks—just tell Jarvis what you need, and watch it navigate your PC, browser, and apps with human-like precision.

✨ Key Capabilities

• 🗣️ Voice & Text Control - Seamlessly switch between typing and talking to your computer.
• 👁️ Visual Intelligence - Powered by advanced Vision-Language Models (VLM) for screen understanding.
• 🖱️ Native GUI Automation - Precise mouse and keyboard control across all applications.
• 🌐 Hybrid Browser Agent - Advanced web navigation using DOM and visual grounding.
• 🧰 MCP Ecosystem - Extensible through Model Context Protocol (MCP) tool integration.
• 🔐 Privacy First - Secure processing with support for local and private models.

---

📺 Showcase

Complex Flight Booking

"Help me book the earliest flight from San Jose to New York on September 1st and the last return flight on September 6th on Priceline"

agent-tars-new-flight.mp4

Action	Local Operator	Remote Operator
VS Code Automation	computer-use-triple-speed.mp4	remote-computer-operators.mp4
GitHub Exploration	browser-use-triple-speed.mp4	remote-browser-operators.mp4

---

🛠️ Quick Start

For Developers (CLI)

Get up and running in seconds with our high-performance CLI.

# Launch instantly with npx
npx @agent-tars/cli@latest

# Or install globally (Requires Node.js >= 22)
npm install @agent-tars/cli@latest -g

# Run Jarvis with your preferred provider
agent-tars --provider anthropic --model claude-3-7-sonnet-latest --apiKey YOUR_API_KEY

Jarvis Desktop App

For a full native experience with a sleek UI:

1. Clone the repository.
2. Run pnpm install.
3. Start the dev environment: npm run dev:ui-tars.
4. Follow the Desktop Quick Start Guide for deeper configuration.

---

📚 Documentation & Resources

Resource	Link
🏠 Website	TBD
📖 Guides	Documentation
🛠️ SDK	Build on Jarvis
🎮 Showcase	Use Cases & Examples

---

🤝 Contributing

We welcome contributions from the community! Whether it's a bug fix, a new feature, or better documentation, check out our CONTRIBUTING.md to get started.

---

📜 License & Credits

Jarvis is open-source software licensed under the Apache License 2.0.

Note

This project is based on and was originally forked from UI-TARS-desktop.

---

_{Built with ❤️ for the future of Human-Computer Interaction.}