seanshin0214/mcp-rag-template

MCP RAG Template

A template for building Model Context Protocol (MCP) servers with RAG (Retrieval-Augmented Generation) capabilities using ChromaDB.

Features

• MCP Server: Full implementation with tools and resources
• RAG with ChromaDB: Vector search for knowledge retrieval
• Skills System: Markdown-based persona and expertise definition
• TypeScript: Type-safe implementation
• Easy Customization: Well-organized structure for your domain

Architecture

┌─────────────────────────────────────────────────────────────┐
│                     MCP RAG Template                        │
├─────────────────────────────────────────────────────────────┤
│                                                             │
│   Skills (Hot Layer)      Tools (Interface)     Knowledge   │
│   ┌─────────────┐        ┌─────────────┐       (RAG Layer)  │
│   │ 01_IDENTITY │        │search_knowl.│       ┌─────────┐  │
│   │ 02_DOMAIN   │   →    │list_collect.│   →   │ChromaDB │  │
│   │ 03_METHODS  │        │get_guide    │       │Vector DB│  │
│   └─────────────┘        └─────────────┘       └─────────┘  │
│                                                             │
└─────────────────────────────────────────────────────────────┘

Quick Start

1. Clone and Setup

git clone https://github.com/seanshin0214/mcp-rag-template.git
cd mcp-rag-template

# Install dependencies
npm install

# Copy environment file
cp .env.example .env

# Build
npm run build

2. Add Your Knowledge

1. Add markdown files to knowledge/ directories
2. Add skill files to skills/ directory
3. Run embedding: npm run embed

3. Configure Claude Desktop

Add to claude_desktop_config.json:

Windows:

{
  "mcpServers": {
    "my-assistant": {
      "command": "node",
      "args": ["C:\\path\\to\\mcp-rag-template\\dist\\index.js"],
      "env": {
        "CHROMA_PATH": "C:\\path\\to\\mcp-rag-template\\chroma-data"
      }
    }
  }
}

macOS/Linux:

{
  "mcpServers": {
    "my-assistant": {
      "command": "node",
      "args": ["/path/to/mcp-rag-template/dist/index.js"],
      "env": {
        "CHROMA_PATH": "/path/to/mcp-rag-template/chroma-data"
      }
    }
  }
}

Project Structure

mcp-rag-template/
├── src/
│   ├── index.ts          # MCP server entry point
│   ├── db/
│   │   └── chromadb.ts   # ChromaDB client
│   ├── tools/
│   │   └── index.ts      # Tool definitions & handlers
│   ├── utils/
│   │   └── skills.ts     # Skills resource handler
│   └── scripts/
│       └── embed.ts      # Knowledge embedding script
├── skills/               # AI persona & expertise (Resources)
│   └── 01_IDENTITY.md
├── knowledge/            # Domain knowledge (for RAG)
│   ├── general/
│   └── guides/
├── package.json
├── tsconfig.json
└── .env.example

Customization Guide

Adding Tools

Edit src/tools/index.ts:

// 1. Add tool definition
export const toolDefinitions = [
  // ... existing tools
  {
    name: "my_new_tool",
    description: "Description of what this tool does",
    inputSchema: {
      type: "object",
      properties: {
        param1: { type: "string", description: "Parameter description" }
      },
      required: ["param1"]
    }
  }
];

// 2. Add handler in handleToolCall switch
case "my_new_tool":
  return handleMyNewTool(args);

// 3. Implement handler
async function handleMyNewTool(args: Record<string, unknown>) {
  // Your implementation
}

Adding Skills

Create markdown files in skills/ directory:

# My Domain Expertise

## Overview
Define your AI's expertise here...

## Capabilities
- Capability 1
- Capability 2

## Guidelines
How the AI should respond...

Adding Knowledge

1. Create markdown files in knowledge/ subdirectories
2. Update EMBED_CONFIG in src/scripts/embed.ts
3. Run npm run embed

Available Scripts

Examples

QualMaster (Qualitative Research)

https://github.com/seanshin0214/qualmaster-mcp

QuantMaster (Quantitative Research)

https://github.com/seanshin0214/quantmaster-mcp

License

MIT License - See for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Built with Model Context Protocol and ChromaDB