Vespucc.ai
Vespucci.ai serves as a bridge between users and AI capabilities provided by MCP servers. The platform uses Groq's language models to process user queries and interact with tools provided by the connected MCP server.
GitHubスター
0
ユーザー評価
未評価
フォーク
0
イシュー
0
閲覧数
1
お気に入り
0
Vespucci.ai Platform
Vespucci.ai is a versatile AI platform that connects to various MCP (Machine Conversation Protocol) servers, enabling powerful AI interactions through multiple interfaces including a REST API and Discord bot integration.
Overview
Vespucci.ai serves as a bridge between users and AI capabilities provided by MCP servers. The platform uses Groq's language models to process user queries and interact with tools provided by the connected MCP server.
Key Features
- MCP Server Integration: Connect to any compatible MCP server to access its tools and capabilities
- Groq LLM Integration: Leverages Groq's powerful language models for natural language processing
- Discord Bot: Interact with the AI through Discord channels and direct messages
- REST API: Programmatically access AI capabilities through a FastAPI backend
- Tool Orchestration: Dynamically discovers and utilizes tools provided by the connected MCP server
- Conversation Logging: Maintains logs of all conversations for reference and debugging
Architecture
Vespucci.ai follows a modular architecture:
- MCP Client: Core component that connects to MCP servers and manages tool interactions
- Language Model Client: Interfaces with Groq's API for LLM capabilities
- Discord Bot: Provides a chat interface through Discord
- FastAPI Server: Exposes REST endpoints for programmatic access
- Configuration System: Manages settings and environment variables
Prerequisites
- Python 3.10 or higher
- Node.js v16 or higher (required for MCP servers)
- Groq API key
- Discord Bot Token (optional, for Discord integration)
Installation
Using Poetry (Recommended)
# Clone the repository
git clone https://github.com/SteeleBlockchainTechnology/Vespucci.ai.git
cd vespucci.ai/mcp
# Install dependencies using Poetry
pip install poetry
poetry install
Environment Setup
Create a .env
file in the project root with the following variables:
GROQ_API_KEY=your_groq_api_key_here
GROQ_MODEL=llama-3-8b-8192 # Optional, this is the default
DISCORD_TOKEN=your_discord_bot_token_here # Optional, for Discord integration
A sample .env.example
file is provided for reference.
Usage
Starting the Platform
# Using Poetry
poetry run python main.py
# Or directly with Python if not using Poetry
python main.py
This will:
- Start the FastAPI server
- Connect to the configured MCP server
- Initialize the Discord bot (if token is provided)
Connecting to Different MCP Servers
By default, Vespucci.ai connects to the web3-research-mcp
server. You can change this in the config/settings.py
file or by setting environment variables.
# In config/settings.py
mcp_args: list = ["--yes", "your-preferred-mcp-server@latest"]
Using the Discord Bot
Once the platform is running with a valid Discord token:
- Invite the bot to your server using the OAuth2 URL from the Discord Developer Portal
- Interact with the bot using the configured command prefix (default:
!
) - Example:
!ask What is the current price of Ethereum?
Using the REST API
The platform exposes several endpoints:
POST /api/query
- Send a query to the AI{ "query": "What is the current price of Ethereum?" }
GET /api/tools
- List available tools from the connected MCP server
Extending Vespucci.ai
Creating Custom MCP Servers
Vespucci.ai can connect to any MCP-compatible server. To create your own:
- Follow the MCP protocol specification
- Implement the required tool interfaces
- Publish your server as an npm package or provide a direct path to your script
Adding New Features
The modular architecture makes it easy to extend:
- Add new API routes in
api/routes.py
- Extend Discord bot commands in
discord_bot/commands.py
- Modify LLM behavior in
client/language_model.py
Conversation Logging
All conversations are logged to the conversations/
directory in JSON format, with timestamps for easy reference.
Troubleshooting
Common Issues
- MCP Server Connection Failure: Ensure Node.js is installed and the MCP server package is accessible
- Groq API Errors: Verify your API key is correct and has sufficient quota
- Discord Bot Not Responding: Check that the token is valid and the bot has proper permissions
License
[Specify your license here]
Acknowledgements
- Groq for providing the LLM API
- MCP Protocol for the tool orchestration framework
- Discord.py for Discord integration
- FastAPI for the REST API framework