ai-embedding-mcp-server

AI Embedding MCP Server

A Model Context Protocol (MCP) server that provides AI embedding and RAG (Retrieval-Augmented Generation) functionality. This server converts your existing AI embedding application into an MCP-compatible service.

Features

• Document Ingestion: Ingest documents and create vector embeddings
• RAG Chat: Chat with AI using business rules context from ingested documents
• Vector Search: Search for similar documents using vector similarity
• Embedding Generation: Generate embeddings for text using HuggingFace models
• Project Management: Organize documents by project ID
• Multiple LLM Support: Works with Ollama (local) and HuggingFace models

Prerequisites

• Node.js 18+
• PostgreSQL with pgvector extension
• Ollama (for local LLM) or HuggingFace API key
• Python 3.8+ (for some dependencies)

Installation

• Clone and install dependencies:

  npm install
  

• Set up PostgreSQL with pgvector:

  # Install pgvector extension
  sudo -u postgres psql -c "CREATE EXTENSION vector;"
  
  # Run the setup script
  psql -U your_username -d your_database -f setup.sql
  

• Configure environment:

  cp env.example .env
  # Edit .env with your configuration
  

• Install Ollama (optional, for local LLM):

  # Install Ollama
  curl -fsSL https://ollama.ai/install.sh | sh
  
  # Pull a model
  ollama pull mistral
  ollama pull qwen2.5-coder:7b
  ollama pull nomic-embed-text
  

Configuration

Edit your .env file with the following variables:

# Database Configuration
DATABASE_URL=postgresql://username:password@localhost:5432/embedding_db

# HuggingFace API Configuration
HUGGINGFACE_API_KEY=your_huggingface_api_key_here

# OpenAI API Configuration (optional)
OPENAI_API_KEY=your_openai_api_key_here

# Ollama Configuration (for local LLM)
OLLAMA_BASE_URL=http://localhost:11434

# Server Configuration
PORT=3000
NODE_ENV=development

Usage

Starting the MCP Server

npm start

The server will run on stdio and can be connected to MCP clients.

Available Tools

1. ingest_document

Ingest a document into the vector database.

Parameters:

• filePath (string): Path to the document file
• projectId (string): Project ID to associate with

Example:

{
  "name": "ingest_document",
  "arguments": {
    "filePath": "./docs/business-rules.mdc",
    "projectId": "project1"
  }
}

2. chat_with_rules

Chat with AI using business rules context.

Parameters:

• message (string): User's question or message
• projectId (string): Project ID to search for context

Example:

{
  "name": "chat_with_rules",
  "arguments": {
    "message": "What are the business rules for the resource module?",
    "projectId": "project1"
  }
}

3. generate_embedding

Generate embeddings for text.

Parameters:

• text (string): Text to generate embeddings for

Example:

{
  "name": "generate_embedding",
  "arguments": {
    "text": "This is sample text for embedding"
  }
}

4. vector_search

Search for similar documents using vector similarity.

Parameters:

• query (string): Search query text
• projectId (string): Project ID to search within
• topK (number, optional): Number of results (default: 5)

Example:

{
  "name": "vector_search",
  "arguments": {
    "query": "business rules",
    "projectId": "project1",
    "topK": 3
  }
}

5. list_projects

List all projects with their document counts.

Example:

{
  "name": "list_projects",
  "arguments": {}
}

6. get_project_documents

Get all documents for a specific project.

Parameters:

• projectId (string): Project ID to get documents for

Example:

{
  "name": "get_project_documents",
  "arguments": {
    "projectId": "project1"
  }
}

Integration with MCP Clients

Claude Desktop

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "ai-embedding": {
      "command": "node",
      "args": ["/path/to/your/ai-embedding/server.js"],
      "env": {
        "DATABASE_URL": "postgresql://username:password@localhost:5432/embedding_db",
        "HUGGINGFACE_API_KEY": "your_api_key"
      }
    }
  }
}

Other MCP Clients

The server follows the MCP protocol and can be integrated with any MCP-compatible client by running:

node server.js

Architecture

The MCP server is built on top of your existing codebase:

• server.js: Main MCP server with tool definitions
• rag-ollama-embedding.js: RAG functionality with Ollama integration
• rag-emedding.js: HuggingFace embedding functionality
• emedding-docs.js: Document processing utilities

Supported File Types

• Text files (.txt, .md, .mdc)
• PDF files (.pdf)
• Word documents (.docx)

Database Schema

The server uses a PostgreSQL database with the following schema:

CREATE TABLE documents (
    id SERIAL PRIMARY KEY,
    project_id VARCHAR(255) NOT NULL,
    file_name VARCHAR(500) NOT NULL,
    content TEXT NOT NULL,
    embedding vector(384),
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

Troubleshooting

Common Issues

• Database Connection Error: Ensure PostgreSQL is running and the connection string is correct
• pgvector Extension Missing: Install the pgvector extension in your PostgreSQL database
• Ollama Connection Error: Ensure Ollama is running and the models are pulled
• HuggingFace API Error: Check your API key and rate limits

Debug Mode

Run with debug logging:

DEBUG=* npm start

Development

Adding New Tools

To add new tools to the MCP server:

• Add the tool definition to the ListToolsRequestSchema handler
• Add the tool implementation to the CallToolRequestSchema handler
• Update this README with the new tool documentation

Testing

Test individual components:

# Test document ingestion
node -e "require('./rag-ollama-embedding.js').ingestDocument('./docs/business-rules.mdc', 'test')"

# Test chat functionality
node -e "require('./rag-ollama-embedding.js').chatWithRules('test message', 'test')"

License

ISC License - see LICENSE file for details.

Contributing

• Fork the repository
• Create a feature branch
• Make your changes
• Test thoroughly
• Submit a pull request

Support

For issues and questions:

• Check the troubleshooting section
• Review the logs for error messages
• Ensure all dependencies are properly installed
• Verify your environment configuration