README.md

metadata

title: 🔄 LLM Conversation Transfer Tool
emoji: 🔄
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.33.1
python_version: 3.1
app_file: app.py
pinned: false
tags:
  - mcp-server-track
  - llm
  - conversation-transfer
  - gradio
  - mcp

🔄 LLM Conversation Transfer Tool

Tags: mcp-server-track

A powerful tool that seamlessly transfers conversations between different LLM providers (ChatGPT, Claude, Mistral, etc.) and functions as both a Gradio web app and an MCP (Model Context Protocol) server.

Link to the Demo video

🌟 Features

• Universal Conversation Parser: Supports JSON format, plain text, and various chat export formats
• Multiple LLM Providers: Transfer to Anthropic Claude, Mistral AI, and Hyperbolic Labs
• Dual Interface: Web app for interactive use + MCP server for programmatic access
• Smart Context Preservation: Maintains conversation flow and context during transfers
• Real-time Status: Live API key validation and connection status

🚀 How to Use

Web Interface

1. Paste Your Conversation: Copy from ChatGPT, Claude, or any chat interface
2. Select Providers: Choose source and target LLM providers
3. Transfer: Click the button and get a response from your target LLM

As MCP Server

This app can be used as an MCP server with any MCP-compatible client:

Available Tool: transfer_conversation

• history_text: Conversation in JSON or plain text format
• source_provider: Source LLM name (ChatGPT, Claude, etc.)
• target_provider: Target LLM (anthropic, mistral, hyperbolic)

📖 MCP Server Demo Video

🎥 Watch the MCP Server in Action

The video demonstrates:

• Setting up the MCP server with Claude Desktop
• Transferring a conversation from ChatGPT to Claude
• Using the tool within an MCP client environment
• Real-time conversation continuation across different LLMs

🔧 Supported Formats

Input Formats

Plain Text:
User: Hello there!
Assistant: Hi! How can I help you?

JSON:
[
  {"role": "user", "content": "Hello there!"},
  {"role": "assistant", "content": "Hi! How can I help you?"}
]

Supported Providers

• ✅ Anthropic (Claude 3 Haiku)
• ✅ Mistral AI (Mistral Small)
• ✅ Hyperbolic Labs (Llama 2)

🛠️ Technical Details

Built with:

• Gradio: Interactive web interface
• MCP (Model Context Protocol): Server functionality
• HTTPX: Async HTTP requests
• Modal: Cloud deployment platform

API Integration

• Anthropic Messages API
• Mistral Chat Completions API
• Hyperbolic Labs API

🚦 Setup & Configuration

Environment Variables

ANTHROPIC_API_KEY=your_anthropic_key
MISTRAL_API_KEY=your_mistral_key
HYPERBOLIC_API_KEY=your_hyperbolic_key

Local Development

# Install dependencies
pip install -r requirements.txt

# Run as web app
python main.py

# Run as MCP server
python mcp_server.py

Modal Deployment

modal deploy main.py

🎯 Use Cases

• LLM Comparison: Test how different models respond to the same conversation
• Context Migration: Move conversations between different AI assistants
• Model Evaluation: Compare responses across multiple LLM providers
• Workflow Integration: Embed in larger AI workflows via MCP protocol

📊 Status Dashboard

The app includes real-time status monitoring:

• API key validation
• Connection health checks
• Transfer success rates
• Error reporting

🔒 Privacy & Security

• No conversation data is stored
• API keys are handled securely through environment variables
• All transfers happen in real-time without logging
• HTTPS connections for all API calls

🤝 Contributing

This project is part of the MCP Server Track. Contributions welcome!

1. Fork the repository
2. Create a feature branch
3. Submit a pull request

📜 License

MIT License - feel free to use and modify!

---

Made for the MCP Server Track 🏆

Seamlessly bridging conversations across the AI ecosystem