QDrant Loader

Name: QDrant Loader
Author: martin-papy

@martin-papy

QDrant Loader について

Enterprise-ready vector database toolkit for building searchable knowledge bases from multiple data sources. Supports multi-project management, automatic ingestion from Confluence/JIRA/Git, intelligent file conversion (PDF/Office/images), and semantic search. Includes MCP server

#python #openai #developer-tools #knowledge-base #file-conversion #enterprise-ready #semantic-search #multi-project #cli-tool #document-processing #embbedings #git-integration #rag #jira-integration #cursor-ide #llm-integration #mcp-server #confluence-integration

基本情報

カテゴリ

データベース

ライセンス

GPL-3.0

ランタイム

python

トランスポート

stdio

公開者

martin-papy

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": {
    "qdrant-loader": {
      "command": "uv",
      "args": [
        "sync",
        "--all-packages",
        "--all-extras"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is QDrant Loader?

QDrant Loader is a data ingestion and retrieval system that collects content from multiple sources, processes and vectorizes it, then provides intelligent search capabilities through a Model Context Protocol (MCP) server for AI-powered development workflows.

How to use QDrant Loader?

Install via pip install qdrant-loader qdrant-loader-mcp-server, initialize a workspace with qdrant-loader init, configure data sources in config.yaml and environment variables in .env, ingest data with qdrant-loader ingest, then start the MCP server using the command mcp-qdrant-loader --env /path/to/.env.

Key features of QDrant Loader

Multi-source connectors: Git, Confluence, JIRA, Public Docs, Local Files.
File conversion: PDF, Office docs, images, audio, EPUB, and more.
Smart chunking with hierarchical context and incremental updates.
Provider-agnostic LLM support: OpenAI, Azure OpenAI, Ollama, custom endpoints.
MCP protocol 2025-06-18 with dual transport (stdio + HTTP).
Advanced search tools: semantic, hierarchy-aware, similarity, clustering, knowledge graphs.

Use cases of QDrant Loader

AI-powered development with Cursor, Windsurf, and other MCP-compatible tools.
Knowledge base creation from technical documentation.
Intelligent code assistance with contextual information.
Enterprise content integration from multiple data sources.

FAQ from QDrant Loader

What does QDrant Loader integrate with?

QDrant Loader integrates with Qdrant vector database, MCP-compatible tools (Cursor, Windsurf, Claude Desktop), and multiple LLM providers (OpenAI, Azure OpenAI, Ollama, custom endpoints).

What are the dependencies and runtime requirements?

QDrant Loader requires Python, a running Qdrant instance, and an LLM provider endpoint. Environment variables for Qdrant URL, collection name, and LLM API key must be configured in a .env file.

Where does ingested data live?

Data is stored in a Qdrant vector database collection specified by the QDRANT_COLLECTION_NAME environment variable. Configuration files (config.yaml) define source connections and processing settings.

What transports and authentication does the MCP server support?

The MCP server supports stdio and HTTP transport (with SSE for streaming). Authentication is handled via environment variables (e.g., OPENAI_API_KEY) and configuration files, with no built-in auth for the HTTP transport mentioned.

How do I configure QDrant Loader for my project?

Create a config.yaml file with project sources (Git, Confluence, JIRA, etc.) and Qdrant/LLM settings, then define environment variables in a .env file for sensitive values. Use qdrant-loader init --workspace . to generate templates.