MCP Qdrant Server With Qdrant Db

🚀 Code Snippet Management Server Based on MCP Protocol Integrated with Qdrant Vector Database

This project combines an MCP (Model Control Protocol) server with a Qdrant vector database to build a system for storing and retrieving code snippets.

🚀 Quick Start

Prerequisites

• Docker container engine
• Docker Compose service orchestration tool

Installation and Running

Step 1: Clone the code repository

git clone <repository-url>
cd <repository-name>

Step 2: Start the service components

docker-compose up -d

Step 3: Check the service status

docker-compose ps

Service access addresses

• Qdrant management interface: http://localhost:6333/dashboard
• MCP Server interface: http://localhost:8000
• For the configuration of AI development environments such as cursor, refer to the following JSON format:

    "qdrant": {
      "url": "http://localhost:8000/sse"
    },

✨ Features

• Save and manage code snippets
• Search for code based on natural language
• Find relevant code through semantic search
• Support the Server-Sent Events (SSE) data transmission protocol

🔧 Technical Details

System Architecture

• Qdrant: Vector database server (Ports: 6333, 6334)
• MCP Server: Code snippet management server (Port: 8000)
• Embedding model: By default, sentence-transformers/all-MiniLM-L6-v2 is used

MCP Toolset

Qdrant MCP Server provides the following functional modules:

1. qdrant-store

• Function description: Save information to the Qdrant database
• Input parameters:
  • information (string): The content of the information to be stored
  • metadata (JSON format): Optional metadata information
• Return result: A confirmation message for successful storage

2. qdrant-find

• Function description: Search for relevant information in the Qdrant database
• Input parameters:
  • query (string): The query statement for performing the search
• Return result: Information stored in the Qdrant database, presented in independent message format

📦 Installation

Environment Variable Configuration

Qdrant-related variables

• RUST_LOG: Set the runtime log level (default value is info)

MCP server-related variables

• QDRANT_URL: Specify the network address of the Qdrant service
• COLLECTION_NAME: Define the data collection name
• EMBEDDING_MODEL: Select the embedding model to use
• TOOL_STORE_DESCRIPTION: Description information for the storage function
• TOOL_FIND_DESCRIPTION: Description information for the search function

Data Persistence

All stored data will be permanently saved in the ./qdrant_storage directory.

Service Management Operations

View running logs

docker-compose logs

Stop the service

docker-compose down

📄 License

No license information provided in the original document, so this section is skipped.

⚠️ Important Note

Regularly back up important data to prevent accidental loss.

💡 Usage Tip

In a production environment, ensure that appropriate security measures are taken.