MCP.so
ログイン

Better Qdrant MCP Server

@wrediam

基本情報

カテゴリ

データベース

ライセンス

MIT

トランスポート

stdio

作者

wrediam

Better Qdrant MCP Server について

概要はまだありません

Better Qdrant MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "better-qdrant-mcp-server": {
      "command": "npx",
      "args": [
        "better-qdrant-mcp-server"
      ]
    }
  }
}

Better Qdrant MCP Server のツール

4

List all available Qdrant collections

Add documents to a Qdrant collection with specified embedding service

Search for similar documents in a collection

Delete a Qdrant collection

概要

What is Better Qdrant MCP Server?

A Model Context Protocol (MCP) server for enhanced Qdrant vector database functionality. It provides tools for managing Qdrant collections, adding documents, and performing semantic searches.

How to use Better Qdrant MCP Server?

Install via npm install -g better-qdrant-mcp-server or run with npx better-qdrant-mcp-server. Configure environment variables (Qdrant URL/keys, embedding service API keys) in a .env file or in your MCP settings for Claude. Use tools like list_collections, add_documents, search, and delete_collection via MCP client.

Key features of Better Qdrant MCP Server

  • List all available Qdrant collections
  • Add documents with multiple embedding services
  • Perform semantic semantic searches across collections
  • Delete collections from the database

Use cases of Better Qdrant MCP Server

  • Manage and organize vector database collections
  • Ingest documents with configurable chunking and embedding
  • Run semantic search queries on stored vectors
  • Clean up unused collections

よくある質問

1

What embedding services are supported?

OpenAI, OpenRouter, Ollama (local, default endpoint http://localhost:11434), and FastEmbed (local models).

2

What are the system requirements?

Node.js >= 18.0.0, a running Qdrant server (local or remote), and API keys for the embedding services you intend to use.

3

How do I configure the server?

Set environment variables in a `.env` file or directly in the MCP client configuration. Required: `QDRANT_URL` and any embedding service API keys. Optional: `QDRANT_API_KEY`, `DEFAULT_EMBEDDING_SERVICE`.

4

How can I use the server with Claude?

Add the server to your MCP settings JSON with `"command": "npx", "args": ["better-qdrant-mcp-server"]` and specify `env` values for Qdrant and embedding services.

5

Can I use local embedding models?

Yes. Ollama (default endpoint http://localhost:11434, model e.g. `nomic-embed-text`) and FastEmbed are supported for local embeddings.

コメント

「データベース」の他のコンテンツ