J.A.R.V.I.S. is an AI-powered coding assistant designed to streamline code generation, modification, and technical discussions for developers. It offers a cross-platform integrated terminal and supports a wide array of state-of-the-art language models, enabling users to select the best AI for their specific coding tasks.

How It Works

J.A.R.V.I.S. utilizes a Flask backend with Flask-SocketIO for real-time communication and Eventlet for asynchronous operations. The frontend is built with pure JavaScript and TailwindCSS, featuring CodeMirror for code editing and Socket.IO client for instant feedback. It supports file attachments including PDFs, DOCX, XLSX, and images (with OCR), processing them via dedicated libraries. The architecture emphasizes real-time updates and workspace management, allowing users to create, rename, and delete multiple coding environments.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Set API keys for various models (DeepSeek, Codestral, OpenAI, Google, Anthropic, etc.) in a .env file.
• Start the server: python app.py
• Access the application at http://localhost:5000.
• Requires Python and standard development tools. API keys for chosen AI models are necessary.

Highlighted Details

• Supports multiple AI models including DeepSeek R1/V3, Codestral, Gemini 2.0 Pro, Grok 2, Claude 3.5 Sonnet, GPT-4 Turbo, GPT-4o, and experimental models.
• Integrated terminal with full xterm.js integration, real-time output streaming, and native shell integration.
• File attachment support with text extraction, OCR for images, and preview capabilities.
• Real-time workspace updates and notifications via WebSockets.

Maintenance & Community

The project acknowledges contributions from Nikole Cardoso, Guilherme Guirro, and Felipe Santos. Further contributions are welcomed via pull requests.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration with closed-source projects. Tested and confirmed working on Linux, WSL 2, and Windows (native, no admin privileges required).

Limitations & Caveats

Several models are listed as "Experimental" or "Preview," indicating potential instability or incomplete functionality. The project relies on numerous external API keys, which may incur costs and require careful management.