MCP Power - Knowledge Search Server
Enables semantic search across multiple knowledge datasets using FAISS vector embeddings, allowing natural language queries to find relevant documents with fast retrieval.
README
<div align="center"> <img src="assets/logo.svg" alt="MCPower Logo" width="120" height="120">
🔍 MCPower
Semantic Knowledge Search, Simplified
Transform your documents into searchable knowledge bases with FAISS vector embeddings
🚀 Quick Start • 📚 Documentation • 🐛 Report Bug • 💡 Request Feature
</div>
✨ What is MCPower?
MCPower is a Model Context Protocol (MCP) server that provides powerful semantic search over your document collections. Drop in any folder of .txt or .md files, and get instant AI-powered search capabilities through a beautiful web interface or programmatic API.
Perfect for:
- 📚 Documentation sites
- 🗂️ Knowledge bases
- 💬 Chatbot context
- 🔍 Research papers
- 📝 Note collections
🎯 Features at a Glance
<table> <tr> <td width="50%">
🖱️ Drag & Drop Interface
Just drop folders into the web console to create searchable datasets. No CLI commands needed!
⚡ Lightning Fast
FAISS-powered vector search with <500ms response times. Search thousands of documents instantly.
🧠 Semantic Understanding
Uses sentence transformers for intelligent matching beyond keyword search.
</td> <td width="50%">
🔌 MCP Compatible
Works with Claude Desktop, VS Code, Cherry Studio, and any MCP client.
📦 Zero Config
One-click launcher automatically sets up everything. Just run ./launch.sh.
🎨 Beautiful UI
Modern, responsive web console with real-time stats and visual feedback.
</td> </tr> </table>
🚀 Quick Start
<details open> <summary><b>🐧 Linux / 🍎 macOS</b></summary>
# Clone the repository
git clone https://github.com/wspotter/mcpower.git
cd mcpower
# Run the launcher - it does everything!
./launch.sh
The web console opens automatically at http://127.0.0.1:4173 🎉
</details>
<details> <summary><b>🪟 Windows</b></summary>
# Clone the repository
git clone https://github.com/wspotter/mcpower.git
cd mcpower
# Double-click launch.bat or run:
launch.bat
Your browser opens automatically to http://127.0.0.1:4173 🎉
</details>
📸 What You'll See
<div align="center"> <img src="assets/screenshot-placeholder.png" alt="MCPower Web Console" width="800"> <p><i>Drag & drop folders, browse for directories, or type paths manually</i></p> </div>
✨ Features
- Semantic Search: Search knowledge datasets using natural language queries
- Interactive Web Console: Manage datasets with drag-and-drop interface
- Multiple Datasets: Manage and search across multiple knowledge bases
- MCP Compatible: Works with any MCP client (VS Code, Cherry Studio, etc.)
- Fast & Reliable: FAISS-powered vector search with <500ms p95 latency
- Graceful Degradation: Continues working even with invalid datasets
- Comprehensive Logging: Structured JSON logs with detailed diagnostics
🏗️ How It Works
<div align="center">
graph TD
A[📄 Your Documents] -->|Python Indexer| B[🧮 Embeddings]
B -->|FAISS| C[💾 Vector Database]
C -->|TypeScript MCP Server| D[🔌 MCP Protocol]
D --> E1[VS Code Copilot]
D --> E2[Cherry Studio]
D --> E3[Any MCP Client]
style A fill:#e3f2fd
style B fill:#fff3e0
style C fill:#f3e5f5
style D fill:#e8f5e9
style E1 fill:#fce4ec
style E2 fill:#fce4ec
style E3 fill:#fce4ec
</div>
The Magic Behind MCPower
-
📚 Document Processing
- Python reads your documents (txt, md, pdf)
- Splits them into semantic chunks
- Generates embeddings using
sentence-transformers
-
⚡ Fast Vector Search
- FAISS indexes embeddings for lightning-fast similarity search
- Sub-500ms query latency even on large datasets
- Efficient memory usage with optimized index structures
-
🔌 MCP Integration
- TypeScript server exposes MCP tools
- Clients send queries via stdio protocol
- Python bridge handles FAISS operations
- Results returned as JSON with relevance scores
⚙️ Installation
Prerequisites
- Node.js 18+ and npm
- Python 3.10+
- Git
Automatic Setup (Recommended)
git clone https://github.com/wspotter/mcpower.git
cd mcpower
./launch.sh # Does everything automatically!
The launcher will:
- ✅ Create virtual environment
- ✅ Install Python dependencies
- ✅ Install Node.js dependencies
- ✅ Configure environment variables
- ✅ Start the web console
- ✅ Open your browser
Manual Setup
<details> <summary><b>Click to expand manual installation steps</b></summary>
1. Clone the repository
git clone https://github.com/wspotter/mcpower.git
cd mcpower
2. Install Node.js dependencies
npm install
3. Create Python virtual environment
python3 -m venv .venv
4. Install Python dependencies
.venv/bin/pip install typer faiss-cpu sentence-transformers
5. Configure environment
cat > .env << EOF
MCPOWER_PYTHON=$(pwd)/.venv/bin/python
EOF
6. Build and run
npm run build
npm run dev -- --datasets ./datasets
</details>
│ └─────┬─────┘ │
└────────┼────────┘
│ execa (JSON over stdio)
│
┌────────┴────────┐
│ Python Bridge │
│ (Python) │
│ │
│ ┌───────────┐ │
│ │ FAISS │ │ Vector search
│ │ Index │ │
│ └───────────┘ │
│ │
│ ┌───────────┐ │
│ │ Sentence │ │ Query encoding
│ │Transformers│ │
│ └───────────┘ │
└─────────────────┘
## 📦 Installation
### Prerequisites
- **Node.js**: 18.x or higher
- **Python**: 3.10 or higher
- **npm**: 9.x or higher
### From Source
```bash
# Clone the repository
git clone https://github.com/yourusername/mcpower.git
cd mcpower
# Install dependencies
npm install
cd python && pip install -r requirements.txt && cd ..
# Build
npm run build
# Optional: Link globally
npm link
Verify Installation
npm run dev -- --version
# Output: Starting MCP Knowledge Server v0.1.0...
⚙️ Configuration
Command Line Options
npm run dev -- [options]
Options:
--datasets <path>: Path to datasets directory (default:./datasets)--log-level <level>: Log level: debug, info, warn, error (default:info)--version: Show version information
Environment Variables
Create a .env file in the project root:
# Datasets directory path
DATASETS_PATH=./datasets
# Log level (debug, info, warn, error)
LOG_LEVEL=info
📚 Dataset Management
Using the Web Console
The easiest way to create datasets is through the web console:
- Start the console:
./launch.sh - Add a dataset:
- Click Browse to open directory picker
- Or drag & drop a folder into the input field
- Or type the path manually
- Submit: Click "Create Dataset"
- Monitor: Watch real-time indexing progress
Dataset Structure
Each dataset has three components stored in datasets/<name>/:
datasets/
└── my-docs/
├── config.json # Dataset configuration
├── index.faiss # FAISS vector index
└── metadata.json # Chunk metadata and text
Manual Dataset Creation
<details> <summary><b>Advanced: Create datasets via Python CLI</b></summary>
# Index a directory of documents
.venv/bin/python python/src/index.py index \
--source-path ./my-documents \
--dataset-name my-docs \
--output-dir ./datasets/my-docs
# Supported file types: .txt, .md, .pdf
Configuration options:
--chunk-size 512 # Characters per chunk
--chunk-overlap 50 # Overlap between chunks
--model sentence-transformers/all-MiniLM-L6-v2
</details>
Dataset Operations
# List all datasets
GET /api/datasets
# Get dataset details
GET /api/datasets/:name
# Delete dataset
DELETE /api/datasets/:name
# Create dataset (via web console or API)
POST /api/datasets
{
"name": "my-docs",
"sourcePath": "/absolute/path/to/documents"
}
└── your-dataset/ ├── manifest.json # Configuration ├── metadata.json # Document metadata └── index/ # FAISS index directory └── docs.index # FAISS index file
### Manifest Format
`manifest.json`:
```json
{
"id": "your-dataset",
"name": "Your Dataset Name",
"description": "Description of your dataset",
"index": "index",
"metadata": "metadata.json",
"defaultTopK": 5
}
Metadata Format
metadata.json:
[
{
"id": "doc-1",
"title": "Document Title",
"path": "path/to/document.md",
"content": "Full document content...",
"snippet": "Short excerpt..."
}
]
Creating a Dataset
See the sample dataset for a complete example.
To create embeddings from your documents, use a tool like:
- sentence-transformers
- FAISS for indexing
- Custom embedding pipeline
� MCP Integration
MCPower works with any MCP-compatible client. Here's how to connect it:
VS Code Copilot
Add to your VS Code settings.json:
{
"github.copilot.chat.codeGeneration.instructions": [
{
"text": "Use the mcpower MCP server for knowledge search"
}
],
"mcp.servers": {
"mcpower": {
"command": "node",
"args": ["/absolute/path/to/mcpower/dist/cli.js", "--datasets", "./datasets"],
"env": {
"MCPOWER_PYTHON": "/absolute/path/to/mcpower/.venv/bin/python"
}
}
}
}
Cherry Studio
Add to Cherry Studio's MCP configuration:
{
"mcpServers": {
"mcpower": {
"command": "node",
"args": ["/absolute/path/to/mcpower/dist/cli.js", "--datasets", "./datasets"]
}
}
}
Available Tools
🔍 knowledge.search
Search your knowledge bases using natural language.
{
dataset: string; // Dataset name (required)
query: string; // Your search query (required)
topK?: number; // Number of results (default: 5)
}
Example:
{
"tool": "knowledge.search",
"arguments": {
"dataset": "my-docs",
"query": "How do I configure authentication?",
"topK": 3
}
}
Response:
{
"results": [
{
"score": 0.89,
"title": "Authentication Guide",
"path": "docs/auth.md",
"snippet": "To configure authentication, set the AUTH_ENABLED=true..."
}
]
}
📋 knowledge.listDatasets
List all available datasets.
{} // No parameters
Response:
{
"datasets": [
{
"id": "my-docs",
"name": "My Documentation",
"description": "Internal docs",
"chunkCount": 1264,
"defaultTopK": 5
}
],
"metadata": {
"total": 1,
"ready": 1,
"errors": 0
}
}
🛠️ Development
Project Structure
mcpower/
├── src/ # TypeScript MCP server
│ ├── cli.ts # Entry point
│ ├── server.ts # MCP protocol handler
│ ├── bridge/ # Python FAISS bridge
│ ├── config/ # Dataset registry
│ ├── store/ # Knowledge store cache
│ └── tools/ # MCP tool implementations
├── python/src/ # Python indexer & search
│ ├── index.py # CLI for indexing
│ └── search.py # FAISS search operations
├── webapp/ # Web console
│ ├── index.html # SPA interface
│ ├── app.js # Frontend logic
│ └── styles.css # Styling
├── tests/ # Test suites
│ ├── unit/ # Unit tests
│ └── integration/ # Integration tests
└── datasets/ # Your knowledge bases
└── sample-docs/ # Example dataset
Development Scripts
# Development mode (auto-reload)
npm run dev -- --datasets ./datasets
# Build TypeScript
npm run build
# Start web console
npm run web
# Run tests
npm test
# Run with coverage
npm run test:coverage
# Type checking & linting
npm run lint
Creating a New Tool
- Define the tool in
src/tools/yourTool.ts:
export const yourTool: Tool = {
name: "knowledge.yourTool",
description: "What your tool does",
inputSchema: {
type: "object",
properties: {
param: { type: "string", description: "Parameter description" }
},
required: ["param"]
}
};
-
Implement the handler in
src/tools/handlers/yourTool.ts -
Register it in
src/server.ts -
Add tests in
tests/unit/tools/yourTool.test.ts
🧪 Testing
Quick Test
# Run all 86 tests
npm test
# Run with coverage report
npm run test:coverage
Test Coverage
✅ 86 tests passing across:
- 🔍 18 search edge cases (empty queries, special chars, large results)
- 🛠️ 15 search tool validations
- 📚 11 dataset registry operations
- 📋 9 listDatasets tool tests
- 🚀 9 startup integration tests
- 💾 8 knowledge store caching
- ⚡ 6 performance benchmarks (<500ms p95)
- 🔗 5+5 integration tests (search + listDatasets)
Integration Testing
# Test with real datasets
./test-search.sh
# Test web console API
./test-web.sh
🐛 Troubleshooting
<details> <summary><b>❌ Dataset Not Found</b></summary>
Error: Dataset not found: your-dataset
Solutions:
- ✅ Verify dataset exists in
datasets/directory - ✅ Check
config.jsonhas correct name field - ✅ Restart server to reload dataset registry
- ✅ Use web console to verify dataset list
</details>
<details> <summary><b>🐍 Python Bridge Failures</b></summary>
Error: Python bridge command failed
Solutions:
- ✅ Verify Python 3.10+ is installed:
python3 --version - ✅ Check virtual environment:
.venv/bin/python --version - ✅ Reinstall dependencies:
.venv/bin/pip install -r python/requirements.txt - ✅ Test FAISS:
.venv/bin/python -c "import faiss; print('OK')" - ✅ Check .env file has correct
MCPOWER_PYTHONpath
</details>
<details> <summary><b>🐌 Slow Search Performance</b></summary>
Issue: Queries taking >500ms
Solutions:
- ✅ Check dataset size (>10k chunks may need optimization)
- ✅ Verify FAISS index is properly trained
- ✅ Reduce
topKparameter (try 3-5 instead of 10+) - ✅ Consider using faster embedding model
- ✅ Use GPU-accelerated FAISS for large datasets
</details>
<details> <summary><b>🌐 Web Console Connection Issues</b></summary>
Error: ERR_CONNECTION_REFUSED
Solutions:
- ✅ Ensure web server is running:
npm run web - ✅ Check port 4173 isn't blocked by firewall
- ✅ Try accessing
http://127.0.0.1:4173directly - ✅ Check console logs for startup errors
</details>
<details> <summary><b>📝 Enable Debug Logging</b></summary>
Get detailed diagnostics:
npm run dev -- --log-level=debug --datasets ./datasets
This shows:
- Dataset loading details
- Python bridge communication
- FAISS index operations
- Search query execution
- Error stack traces
</details>
🤝 Contributing
We welcome contributions! Here's how to get started:
Quick Start
# Fork and clone
git clone https://github.com/YOUR_USERNAME/mcpower.git
cd mcpower
# Create feature branch
git checkout -b feature/amazing-feature
# Install dependencies
npm install
.venv/bin/pip install -r python/requirements.txt
# Make changes and test
npm run build
npm test
# Commit with clear message
git commit -m "feat: add amazing feature"
# Push and create PR
git push origin feature/amazing-feature
Areas We Need Help
- 🎨 UI/UX: Improve web console design
- 📚 Documentation: Tutorials, examples, guides
- 🧪 Testing: More test coverage, edge cases
- 🚀 Performance: Optimization, caching strategies
- 🔌 Integrations: New MCP clients, data sources
- 🐛 Bug Fixes: See issues
Code Guidelines
- Write tests for new features
- Follow TypeScript/Python best practices
- Update documentation for API changes
- Use conventional commit messages
- Keep PRs focused and atomic
📄 License
MIT License - see LICENSE for details
🙏 Acknowledgments
Built with amazing open-source tools:
- FAISS - Vector similarity search by Facebook Research
- sentence-transformers - State-of-the-art text embeddings
- MCP - Model Context Protocol by Anthropic
- TypeScript - Type-safe JavaScript
- Express - Fast web framework
<div align="center">
⭐ Star this repo if you find it useful!
Made with ❤️ by the MCPower team
🐛 Report Bug • ✨ Request Feature • 📖 Documentation
</div>
Look for error logs with:
manifestPath: Location of problematic manifesterrorType: Type of error (json_parse_error, file_not_found, validation_error)error: Detailed error message
📝 License
ISC
🤝 Contributing
Contributions welcome! Please read our contributing guidelines before submitting PRs.
🔗 Links
📊 Project Status
- ✅ Phase 1-5: Complete (All user stories implemented)
- 🚧 Phase 6: Polish & documentation (in progress)
Made with ❤️ for the MCP community
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.