This project provides an advanced chat application that leverages a unique debate paradigm between two AI personas to refine user queries and deliver synthesized answers. It's designed for users seeking more robust, accurate, and critically examined AI responses, offering flexibility through multiple backend API support.

How It Works

The core of the application is a dual AI debate system. User input triggers a discussion between a logical AI (Cognito) and a skeptical AI (Muse). This dialectical process aims to reduce AI hallucinations and stress-test information before presenting a final, consolidated answer in a shared, Markdown-editable notepad with undo/redo functionality.

Quick Start & Requirements

• Install: Clone the repository and run npm install.
• Prerequisites: Node.js (v18 or higher).
• API Configuration: Requires a Google Gemini API key or an OpenAI-compatible API endpoint and model IDs. Configuration can be done via .env.local or through the in-app settings.
• Run: npm run dev
• Demo/Docs: Application Screenshot

Highlighted Details

• Supports multimodal input (text and images).
• Offers flexible API backend support for Google Gemini and any OpenAI-compatible API (e.g., Ollama, LM Studio).
• Features configurable discussion modes: fixed rounds or AI-driven.
• Includes robust error handling with automatic and manual retry options.

Maintenance & Community

The project was conceived and created by "yeahhe" on the LINUX DO forum. No specific community links or active maintenance signals are provided in the README.

Licensing & Compatibility

The project is licensed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

The application does not support streaming responses, waiting for the full AI generation before display. In "AI-driven" mode, there's a theoretical possibility of AI debate loops. All AI requests are executed sequentially, meaning users must wait for one step to complete before proceeding to the next.