Easy MCP RAG
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Easy MCP RAG ๐
A high-performance Model Context Protocol (MCP) server for RAG using Qdrant. Built for UV/UVX with CPU/GPU support and HTTP transport.
โจ Features
๐ Automatic Document Indexing - Scan directories and index all documents
๐ Smart Organization - Each subdirectory becomes its own searchable dataset
๐ ๏ธ Dynamic MCP Tools - Auto-generated tools for each collection
๐ Multi-Format Support - PDF, DOCX, CSV, XLSX, TXT, Markdown, and more
โก GPU Acceleration - Optional CUDA/MPS support for faster embeddings
๐ HTTP Transport - Run as HTTP server or stdio
๐ฆ UV/UVX Ready - Install and run with a single command
๐ Verbose Logging - Detailed query tracking and monitoring
Related MCP server: mcp-rag-server
๐ Quick Start
Install with UVX (Recommended)
Run directly from GitHub without installation:
uvx --from git+https://github.com/yourusername/easy_mcp_rag.git easy_mcp_rag --data-dir ./documentsInstall with UV
# Install from GitHub
uv pip install git+https://github.com/yourusername/easy_mcp_rag.git
# Or clone and install locally
git clone https://github.com/yourusername/easy_mcp_rag.git
cd easy_mcp_rag
uv pip install -e .๐ Prerequisites
Start Qdrant (using Docker):
docker run -p 6333:6333 qdrant/qdrantPrepare your documents:
documents/
โโโ legal_docs/
โ โโโ contract.pdf
โ โโโ terms.docx
โโโ research/
โ โโโ paper1.pdf
โ โโโ notes.txt
โโโ data/
โโโ analysis.csv๐ป Usage
Basic Usage (stdio)
# With UVX
uvx --from git+https://github.com/yourusername/easy_mcp_rag.git easy_mcp_rag --data-dir ./documents
# With UV
uv run easy_mcp_rag --data-dir ./documents
# After installation
easy_mcp_rag --data-dir ./documentsHTTP Mode
easy_mcp_rag --data-dir ./documents --transport http --http-port 8000GPU Acceleration
# Auto-detect GPU
easy_mcp_rag --data-dir ./documents --device auto
# Force CUDA (NVIDIA GPU)
easy_mcp_rag --data-dir ./documents --device cuda
# Force MPS (Apple Silicon)
easy_mcp_rag --data-dir ./documents --device mps
# Force CPU
easy_mcp_rag --data-dir ./documents --device cpuAdvanced Configuration
easy_mcp_rag \
--data-dir ./documents \
--qdrant-host localhost \
--qdrant-port 6333 \
--device cuda \
--embedding-model all-mpnet-base-v2 \
--chunk-size 1024 \
--chunk-overlap 100 \
--top-k 10 \
--batch-size 64 \
--verbose \
--force-reindex๐ง Configuration Options
Flag | Description | Default |
| Directory with document subdirectories | Required |
| Qdrant server host |
|
| Qdrant server port |
|
| Device: auto, cpu, cuda, mps |
|
| Transport type: stdio, http |
|
| HTTP server host |
|
| HTTP server port |
|
| Sentence transformer model |
|
| Text chunk size (chars) |
|
| Chunk overlap (chars) |
|
| Results per search |
|
| Embedding batch size |
|
| Enable verbose logging |
|
| Log level |
|
| Force reindex all docs |
|
๐ฏ MCP Client Configuration
Claude Desktop / Cline / Other MCP Clients
Add to your MCP client config:
{
"mcpServers": {
"rag-server": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/yourusername/easy_mcp_rag.git",
"easy_mcp_rag",
"--data-dir",
"/path/to/your/documents",
"--device",
"auto",
"--verbose"
]
}
}
}With HTTP Transport
{
"mcpServers": {
"rag-server": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/yourusername/easy_mcp_rag.git",
"easy_mcp_rag",
"--data-dir",
"/path/to/your/documents",
"--transport",
"http",
"--http-port",
"8000"
]
}
}
}๐ ๏ธ How It Works
Scan - Discovers all subdirectories in your data directory
Load - Extracts text from all supported file types
Chunk - Splits documents into overlapping chunks
Embed - Generates vector embeddings (CPU or GPU)
Index - Stores in Qdrant (one collection per subdirectory)
Serve - Creates MCP tools for each collection
Example
documents/
โโโ legal_docs/ โ Creates "legal_docs_search" tool
โโโ research/ โ Creates "research_search" tool
โโโ data/ โ Creates "data_search" tool๐ Supported File Types
Category | Extensions |
Text |
|
| |
Word |
|
Spreadsheet |
|
๐จ Embedding Models
Choose based on your needs:
Model | Dimensions | Speed | Quality | Use Case |
| 384 | โกโกโก | Good | Default, fast |
| 384 | โกโก | Better | Balanced |
| 768 | โก | Best | Quality |
๐ Troubleshooting
Qdrant Connection Failed
# Check if Qdrant is running
curl http://localhost:6333
# Start Qdrant
docker run -p 6333:6333 qdrant/qdrantGPU Not Detected
# Check PyTorch GPU support
python -c "import torch; print(torch.cuda.is_available())"
# Install with GPU support
uv pip install -e ".[gpu]"Out of Memory
# Use smaller model
--embedding-model all-MiniLM-L6-v2
# Reduce batch size
--batch-size 16
# Use CPU
--device cpu๐ Logging
Enable verbose logging to see detailed information:
easy_mcp_rag --data-dir ./documents --verboseOutput includes:
โ Tool access events
๐ Query details
๐ Result counts
๐ฏ Relevance scores
๐ Source files
Example:
2024-01-20 10:30:15 - easy_mcp_rag.server - INFO - Tool accessed: legal_docs_search
2024-01-20 10:30:15 - easy_mcp_rag.server - INFO - Query: contract terms
2024-01-20 10:30:15 - easy_mcp_rag.server - INFO - Results returned: 5
2024-01-20 10:30:15 - easy_mcp_rag.server - DEBUG - Result 1: score=0.8542๐ Security Notes
HTTP mode exposes the server on the network
Use
--http-host 127.0.0.1for local-only accessConsider authentication for production deployments
๐ Development
# Clone repository
git clone https://github.com/yourusername/easy_mcp_rag.git
cd easy_mcp_rag
# Install with dev dependencies
uv pip install -e ".[dev]"
# Run tests
pytest
# Format code
black src/
# Lint
ruff src/๐ค Contributing
Contributions welcome! Please:
Fork the repository
Create a feature branch
Make your changes
Submit a pull request
๐ License
MIT License - see LICENSE file
๐ Credits
Built with:
MCP - Model Context Protocol
Qdrant - Vector database
Sentence Transformers - Embeddings
UV - Package manager
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogโฆ
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP server for agentverse documentation, generated by doc2mcp.
Driflyte MCP server which lets AI assistants query topic-specific knowledge from web and GitHub.
Cloud or self-hosted knowledge for AI agents: hybrid search, reranking, GraphRAG, scoped MCP tools.
Related MCP Servers
- AlicenseDqualityDmaintenanceA lightweight RAG system that provides an MCP server for searching and interacting with vector-based knowledge bases. It enables users to perform retrieval-augmented generation and search across Qdrant collections through a standardized interface.12MIT
- AlicenseNot gradedqualityDmaintenanceAn MCP server that indexes documents and serves relevant context to LLMs via Retrieval Augmented Generation (RAG).36MIT
- AlicenseBqualityBmaintenanceA production-grade MCP server for integrating RAG into AI agents, supporting multiple vector databases with enterprise security and dynamic tool selection.716MIT
- FlicenseNot gradedqualityDmaintenanceAn enterprise-ready MCP server that exposes a RAG tool for retrieving relevant context and metadata from a Qdrant vector database using natural language queries.2-