Model Context Protocol server implementation for Qdrant vector database.
- Implements MCP protocol for Qdrant vector database
- Supports embedding generation with FastEmbed
- Configurable through environment variables
- Docker support for easy deployment
pip install mcp-server-qdrantSet required environment variables:
export QDRANT_URL=http://localhost:6333
export COLLECTION_NAME=my_collectionRun the server:
mcp-server-qdrantRun tests:
docker-compose -f docker-compose.test.yml up --build-
Install prerequisites:
- Docker Desktop
- Claude Desktop or another MCP client
- Git
-
Clone and start services:
git clone https://github.com/your-org/mcp-server-qdrant
cd mcp-server-qdrant
docker-compose up -d-
Configure your MCP client (see Task Management Guide)
-
Start using the system:
I have a test failure in test_authentication. Can you help me find similar issues?
# Create virtual environment
python -m venv .venv
source .venv/bin/activate # or .venv\Scripts\activate on Windows
# Install dependencies
pip install -r requirements.txt
# Run tests
pytest- Fork the repository
- Create your feature branch
- Commit your changes
- Push to the branch
- Create a Pull Request
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
- Python 3.11 or higher
- Docker (optional, for containerized deployment)
- Node.js and npm (for Claude Desktop integration)
We provide detailed setup guides for different platforms and deployment methods:
-
- For Windows, macOS, and Linux users using Docker Desktop
- Includes container configuration and Claude Desktop integration
- Best for users who prefer containerized deployment
-
- For RHEL 9 and compatible distributions using Podman
- Includes SELinux configuration and security considerations
- Optimized for enterprise Linux environments
-
- For Windows users running directly on the host
- Includes detailed Python and Node.js setup
- Best for local development on Windows
Choose the guide that best matches your platform and deployment preferences.
The server can be configured using environment variables:
| Variable | Description | Default |
|---|---|---|
| QDRANT_URL | URL of the Qdrant server | None |
| QDRANT_LOCAL_PATH | Path to local Qdrant storage | None |
| COLLECTION_NAME | Name of the Qdrant collection | Required |
| EMBEDDING_PROVIDER | Embedding provider (fastembed) | fastembed |
| EMBEDDING_MODEL | Model name for embeddings | sentence-transformers/all-MiniLM-L6-v2 |
Note: You must specify either QDRANT_URL or QDRANT_LOCAL_PATH, but not both.
- Run all tests:
./scripts/e2e_test.sh- Run specific test categories:
# Unit tests only
python -m pytest tests/unit/
# Integration tests only
python -m pytest tests/integration/
# Settings tests
python -m pytest tests/test_settings.pyTo generate a test coverage report:
python -m pytest --cov=src/mcp_server_qdrant tests/hatch builddocker build \
--build-arg VERSION=$(cat VERSION) \
--build-arg BUILD_DATE=$(date -u +'%Y-%m-%dT%H:%M:%SZ') \
--build-arg VCS_REF=$(git rev-parse --short HEAD) \
-t mcp-server-qdrant:latest .# Set required environment variables
export COLLECTION_NAME=my_collection
export QDRANT_URL=http://localhost:6333
# Run the server
python -m mcp_server_qdrantdocker run -d \
-p 8000:8000 \
-e COLLECTION_NAME=my_collection \
-e QDRANT_URL=http://qdrant:6333 \
mcp-server-qdrant:latestFor local development using uvx:
{
"qdrant": {
"command": "uvx",
"args": ["mcp-server-qdrant"],
"env": {
"QDRANT_URL": "https://xyz-example.eu-central.aws.cloud.qdrant.io:6333",
"QDRANT_API_KEY": "your_api_key",
"COLLECTION_NAME": "your-collection-name",
"EMBEDDING_MODEL": "sentence-transformers/all-MiniLM-L6-v2"
}
}
}For Docker deployment:
75E7{
"qdrant": {
"command": "docker",
"args": [
"run",
"--rm",
"--network", "mcp-network",
"-p", "8080:8080",
"-e", "QDRANT_URL=http://qdrant:6333",
"-e", "MCP_COLLECTION=mcp_unified_store",
"-e", "EMBEDDING_MODEL=sentence-transformers/all-MiniLM-L6-v2",
"-e", "LOG_LEVEL=INFO",
"mcp-server-qdrant:latest"
]
}
}For cloud-hosted Qdrant:
{
"qdrant": {
"command": "docker",
"args": [
"run",
"--rm",
"-p", "8080:8080",
"-e", "QDRANT_URL=https://xyz-example.eu-central.aws.cloud.qdrant.io:6333",
"-e", "QDRANT_API_KEY=your_api_key",
"-e", "MCP_COLLECTION=your-collection-name",
"-e", "EMBEDDING_MODEL=sentence-transformers/all-MiniLM-L6-v2",
"-e", "LOG_LEVEL=INFO",
"mcp-server-qdrant:latest"
]
}
}- Fork the repository
- Create a feature branch
- Make your changes
- Run tests
- Submit a pull request
[Your License Here]