https://github.com/bjornmelin/mcp-search-hub
🔍 Intelligent multi-provider search aggregation server built on FastMCP 2.0
https://github.com/bjornmelin/mcp-search-hub
Last synced: 4 months ago
JSON representation
🔍 Intelligent multi-provider search aggregation server built on FastMCP 2.0
- Host: GitHub
- URL: https://github.com/bjornmelin/mcp-search-hub
- Owner: BjornMelin
- License: mit
- Created: 2025-05-14T23:55:08.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2025-06-16T04:13:10.000Z (4 months ago)
- Last Synced: 2025-06-16T05:25:17.952Z (4 months ago)
- Language: Python
- Size: 911 KB
- Stars: 4
- Watchers: 1
- Forks: 0
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# MCP Search Hub
> 🔍 **Intelligent multi-provider search aggregation server built on FastMCP 2.0**
[](https://opensource.org/licenses/MIT)
[](https://www.python.org/downloads/)
[](https://github.com/fastmcp)
MCP Search Hub embeds official MCP servers from five leading search providers within a unified interface, intelligently routes queries to the most appropriate provider(s), and combines/ranks results for optimal relevance.
## ✨ Key Features
- **🏗️ Unified Interface**: Single MCP server for all search providers
- **🧠 Intelligent Routing**: Automatically selects optimal providers based on query characteristics
- **⚡ Official Provider Integration**: Embeds official MCP servers (Linkup, Exa, Perplexity, Tavily, Firecrawl)
- **💰 Cost Optimization**: 30-45% cost reduction through intelligent provider selection
- **🚀 Zero Maintenance**: Provider updates flow through automatically
- **📊 Smart Result Merging**: Deduplication and multi-factor ranking
- **⚡ Performance Optimized**: Multi-tier caching and parallel execution
- **🔧 Production Ready**: Error handling, rate limiting, monitoring
## 🚀 Quick Start
Get up and running in 5 minutes:
```bash
# 1. Clone and install
git clone https://github.com/BjornMelin/mcp-search-hub
cd mcp-search-hub
uv pip install -r requirements.txt
# 2. Interactive setup (recommended)
python scripts/setup_claude_desktop.py
# 3. Start searching!
# The script will guide you through API key setup and Claude Desktop integration
```
After setup, restart Claude Desktop and ask: *"What MCP tools do you have available?"*
**📖 [Complete Setup Guide](docs/users/getting-started.md)** | **⚡ [Quick Setup Script](scripts/setup_claude_desktop.py)**
## 🎯 Why MCP Search Hub?
### Provider Strengths
Each provider excels in different areas. MCP Search Hub automatically routes queries to leverage these strengths:
| Provider | Strengths | Best For | Accuracy |
|----------|-----------|----------|----------|
| **Linkup** | Factual information, current events | News, facts, real-time data | 91.0% |
| **Exa** | Semantic search, academic content | Research papers, technical docs | 90.0% |
| **Perplexity** | AI-powered analysis, current events | Complex queries, analysis | 86.0% |
| **Tavily** | RAG-optimized results | Retrieval applications | 73.0% |
| **Firecrawl** | Deep content extraction | Web scraping, full content | N/A |
*Accuracy scores from SimpleQA benchmark*
### Architecture Benefits
```mermaid
graph TB
User[Claude Desktop/Code] --> Hub[MCP Search Hub]
Hub --> Router[Intelligent Router]
Router -->|Academic Query| Exa[Exa MCP Server]
Router -->|News Query| Perplexity[Perplexity MCP Server]
Router -->|Factual Query| Linkup[Linkup MCP Server]
Router -->|Scraping Task| Firecrawl[Firecrawl MCP Server]
Hub --> Cache[Smart Caching]
Hub --> Merger[Result Merger]
style Hub fill:#e1f5fe
style Router fill:#f3e5f5
style Cache fill:#e8f5e8
style Merger fill:#fff3e0
```
**Benefits:**
- **Single Integration**: One MCP server instead of five
- **Zero Configuration**: No need to manage multiple servers
- **Automatic Updates**: Provider improvements flow through automatically
- **Cost Efficient**: Intelligent routing reduces unnecessary API calls
- **Quality Results**: Multi-provider consensus improves accuracy
## 📚 Documentation
### For Users
- **[Getting Started](docs/users/getting-started.md)** - Complete setup guide and first steps
- **[Configuration](docs/operators/configuration.md)** - Detailed configuration reference
- **[API Reference](docs/developers/api-reference.md)** - Complete tool and endpoint documentation
- **[Troubleshooting](docs/troubleshooting/common-issues.md)** - Solutions to common problems
### For Developers
- **[Contributing](docs/developers/contributing.md)** - How to contribute to the project
- **[Development](docs/developers/development.md)** - Development workflows and practices
- **[Architecture](docs/architecture/overview.md)** - System design and components
### Advanced Topics
- **[Provider Integration](docs/architecture/provider-integration.md)** - How providers are embedded
- **[Caching Strategy](docs/architecture/caching.md)** - Multi-tier caching system
- **[Performance Tuning](docs/operators/production-setup.md)** - Optimization strategies
📖 **[Complete Documentation Index](docs/README.md)**
## 🔑 Quick Configuration
At minimum, you need API keys for the providers you want to use:
```bash
# Copy environment template
cp .env.template .env
# Add your API keys
LINKUP_API_KEY=your_linkup_key
EXA_API_KEY=your_exa_key
PERPLEXITY_API_KEY=your_perplexity_key
TAVILY_API_KEY=your_tavily_key
FIRECRAWL_API_KEY=your_firecrawl_key
```
### Getting API Keys
| Provider | Free Tier | Sign Up |
|----------|-----------|---------|
| **Linkup** | 100 requests/month | [linkup.so](https://linkup.so) |
| **Exa** | 1,000 requests/month | [exa.ai](https://exa.ai) |
| **Perplexity** | $5 credit | [perplexity.ai](https://perplexity.ai) |
| **Tavily** | 1,000 requests/month | [tavily.com](https://tavily.com) |
| **Firecrawl** | 500 requests/month | [firecrawl.dev](https://firecrawl.dev) |
⚙️ **[Complete Configuration Guide](docs/operators/configuration.md)**
## 🚀 Available Tools
MCP Search Hub provides all tools from embedded provider MCP servers:
### Core Search Tools
- **`search`** - Intelligent multi-provider search with automatic routing
- **`get_provider_info`** - Information about available providers
### Provider-Specific Tools (25+ tools available)
- **Firecrawl**: `firecrawl_scrape`, `firecrawl_search`, `firecrawl_crawl`, `firecrawl_extract`
- **Exa**: `web_search_exa`, `research_paper_search`, `company_research`, `linkedin_search`
- **Perplexity**: `perplexity_ask`, `perplexity_research`
- **Linkup**: `linkup_search_web`
- **Tavily**: `tavily_search`, `tavily_extract`
🛠️ **[Full Tool Documentation](docs/developers/api-reference.md)**
## 🏗️ Architecture
MCP Search Hub uses a modular architecture with embedded MCP servers:
```mermaid
graph TB
subgraph "MCP Search Hub"
Server[FastMCP Server]
Router[Query Router]
Cache[Tiered Cache]
Merger[Result Merger]
end
subgraph "Embedded MCP Servers"
LinkupMCP[Linkup MCP]
ExaMCP[Exa MCP]
PerplexityMCP[Perplexity MCP]
TavilyMCP[Tavily MCP]
FirecrawlMCP[Firecrawl MCP]
end
Client[Claude Desktop] --> Server
Server --> Router
Router --> LinkupMCP
Router --> ExaMCP
Router --> PerplexityMCP
Router --> TavilyMCP
Router --> FirecrawlMCP
style Server fill:#e1f5fe
style Router fill:#f3e5f5
style Cache fill:#e8f5e8
style Merger fill:#fff3e0
```
🏗️ **[Architecture Documentation](docs/architecture/overview.md)**
## 🤝 Contributing
We welcome contributions! Here's how to get started:
1. **Fork** the repository
2. **Create** a feature branch
3. **Make** your changes with tests
4. **Submit** a pull request
🤝 **[Contributing Guide](docs/developers/contributing.md)** | 🛠️ **[Development Setup](docs/developers/development.md)**
## 📄 License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## 🙏 Acknowledgements
- **[FastMCP](https://github.com/fastmcp)** - The framework powering this server
- **Provider Teams** - For their excellent APIs and official MCP servers
## 🔗 Links
- **[GitHub Repository](https://github.com/BjornMelin/mcp-search-hub)**
- **[Documentation](docs/README.md)**
- **[Issue Tracker](https://github.com/BjornMelin/mcp-search-hub/issues)**
- **[Releases](https://github.com/BjornMelin/mcp-search-hub/releases)**
---
**Ready to get started?** Follow our [Getting Started Guide](docs/users/getting-started.md) or run the interactive setup:
```bash
python scripts/setup_claude_desktop.py
```