🔍 Presearch MCP Server

Privacy-first search integration for AI assistants - A professional Model Context Protocol (MCP) server that seamlessly integrates Presearch's decentralized search API with AI applications.

🌟 Overview

Presearch MCP Server provides a robust, privacy-focused bridge between AI assistants and the Presearch decentralized search engine. Built with enterprise-grade architecture, it offers intelligent caching, multi-format exports, and comprehensive web scraping capabilities while maintaining user privacy and data security.

🔗 Quick Links

Repository: GitHub
Smithery: Smithery.ai
Presearch API: Developer Portal
Issues: Support

✨ Key Features

🛡️ Privacy & Security

Decentralized Search: Leverages Presearch's distributed network for private, uncensored search results
No Tracking: Zero user data collection or behavioral tracking
Secure Authentication: Bearer token authentication with configurable API endpoints

🔧 AI Integration

MCP Protocol Compliance: Full Model Context Protocol implementation for seamless AI assistant integration
Dual Transport Support: Both HTTP and STDIO transport modes
Smithery.ai Ready: Pre-configured for instant deployment on Smithery platform

🚀 Performance & Reliability

Intelligent Caching: Redis-compatible caching with configurable TTL and key limits
Rate Limiting: Built-in request throttling and retry mechanisms
Error Handling: Comprehensive error handling with detailed logging
Health Monitoring: Built-in health checks and performance metrics

📊 Data Export & Analysis

Multi-Format Support: JSON, CSV, Markdown, HTML, and PDF export formats
Content Scraping: Intelligent web content extraction with metadata preservation
Batch Processing: Bulk search and scraping operations
Quality Scoring: Advanced result ranking and quality assessment

🚀 Quick Start

Prerequisites

Node.js ≥20.0.0
Presearch API key (Get yours here)
Docker (optional, for containerized deployment)

Installation

# Clone the repository git clone https://github.com/NosytLabs/presearch-search-api-mcp.git cd presearch-search-api-mcp # Install dependencies npm install # Configure environment cp .env.example .env

Configuration

Edit .env file with your settings:

# Required: Your Presearch API key PRESEARCH_API_KEY=your_presearch_api_key_here # Optional: API configuration PRESEARCH_BASE_URL=https://na-us-1.presearch.com # Optional: Server configuration PORT=3000 NODE_ENV=production # Optional: Cache settings CACHE_ENABLED=true CACHE_TTL=3600 CACHE_MAX_KEYS=1000 # Optional: Logging LOG_LEVEL=info

Start the Server

HTTP Mode (for remote connections):

npm start

STDIO Mode (for local AI assistants):

npm run start:stdio

Development Mode (with auto-reload):

npm run dev

Docker Mode (for containerized deployment):

docker build -t presearch-mcp-server . docker run -p 3000:3000 -v ./.env:/app/.env presearch-mcp-server

🔌 MCP Configuration

HTTP Transport Configuration

Add to your MCP client configuration:

{ "mcpServers": { "presearch": { "transport": "http", "url": "http://localhost:3000/mcp", "headers": { "Authorization": "Bearer your-api-key" } } } }

STDIO Transport Configuration

{ "mcpServers": { "presearch": { "transport": "stdio", "command": "node", "args": ["src/index.js", "--stdio"], "env": { "PRESEARCH_API_KEY": "your-api-key", "NODE_ENV": "production" } } } }

🛠️ Available Tools

Search Operations

presearch_ai_search - Perform intelligent web searches with advanced filtering
presearch_search_and_scrape - Search and automatically scrape top results
export_search_results - Export search results in multiple formats

Content Extraction

scrape_url_content - Extract content from specified URLs
export_site_content - Export scraped content with formatting
content_analysis - Analyze content quality and relevance

System Management

presearch_health_check - Verify API connectivity and authentication
cache_stats - View cache performance metrics
cache_clear - Clear cached data

📖 Usage Examples

Basic Search

{ "tool": "presearch_ai_search", "params": { "query": "decentralized finance trends 2024", "count": 10, "safesearch": "moderate", "freshness": "week" } }

Advanced Search with Scraping

{ "tool": "presearch_search_and_scrape", "params": { "query": "blockchain technology adoption", "count": 5, "scrape_count": 3, "lang": "en-US", "file_output": true, "export_format": "markdown" } }

Content Scraping

{ "tool": "scrape_url_content", "params": { "urls": [ "https://presearch.io", "https://example.com/article" ], "include_metadata": true, "export_format": "json" } }

Exporting Search Results

{ "tool": "export_search_results", "params": { "query": "latest advancements in AI", "count": 20, "export_format": "csv", "file_path": "/path/to/save/results.csv" } }

🔐 Presearch API Authentication

Obtaining API Keys

Visit Presearch Developer Portal
Create a developer account
Generate your API key (Bearer token)
Add to your .env file: PRESEARCH_API_KEY=your_key_here

Authentication Headers

The server automatically handles authentication using the Authorization: Bearer <token> header format as specified in the Presearch API documentation.

API Endpoints

Base URL: https://na-us-1.presearch.com/v1/search
Authentication: Bearer token in Authorization header
Rate Limits: Configurable per your Presearch plan

🐳 Docker Deployment

Quick Docker Setup

# Build the image docker build -t presearch-mcp . # Run with environment variables docker run -p 3000:3000 -e PRESEARCH_API_KEY=your_key_here presearch-mcp

Docker Compose

version: '3.8' services: presearch-mcp: build: . ports: - "3000:3000" environment: - PRESEARCH_API_KEY=your_key_here - NODE_ENV=production volumes: - ./.env:/app/.env

🔧 Development

Project Structure

presearch-search-api-mcp/ ├── src/ │ ├── index.js # Main server entry │ ├── tools/ # MCP tool implementations │ ├── utils/ # Utility functions │ └── config/ # Configuration files ├── tests/ # Test files ├── docs/ # Documentation └── exports/ # Generated exports

Available Scripts

npm start - Start HTTP server
npm run start:stdio - Start STDIO server
npm run dev - Development mode with auto-reload
npm run lint - Run ESLint
npm run format - Format code with Prettier

Testing

# Run tests npm test # Run with coverage npm run test:coverage

📊 Performance & Monitoring

Cache Metrics

Hit/miss ratios
Response time tracking
Memory usage monitoring

Health Checks

API connectivity verification
Authentication validation
System resource monitoring

🔒 Security Features

Data Protection

No user data persistence
Encrypted API communications
Secure token handling

Rate Limiting

Configurable request limits
Automatic retry mechanisms
DDoS protection

📝 License

MIT License - see LICENSE file for details.

🤝 Contributing

Fork the repository
Create a feature branch
Make your changes
Add tests
Submit a pull request

🐛 Support

Issues: GitHub Issues
Documentation: Wiki
Discussions: GitHub Discussions

🌟 Star History

Built with ❤️ for the privacy-conscious AI community