Clinical Trials MCP Server

Extract and analyze clinical trial data from ClinicalTrials.gov with AI-powered search and summarization.

Installation

npm install
npm run build

Usage

MCP Server

Add to your mcp.json:

{
  "mcpServers": {
    "clinical-trials": {
      "command": "node",
      "args": ["/path/to/clinical-trials/dist/mcp/server.js"]
    }
  }
}

Features

• 🔍 Advanced Search: Search across 19 specialized areas (conditions, interventions, locations, etc.)
• 🎯 Iterative Refinement: Filter results without re-querying the API
• 🤖 AI Integration: MCP server for conversational access through MCP Client
• 💾 Smart Caching: In-memory and disk caching to minimize API calls
• 📊 Flexible Export: CSV, JSON, JSONL formats
• 🗄️ Local Database: SQLite storage with full-text search

Iterative Refinement

The tool creates sessions that allow you to refine searches without hitting the API again:

1. Initial search hits API and stores results
2. Filter operations work on cached results
3. Session persists in database for future access

Smart Caching

• Memory cache: 1 minute TTL for instant repeated queries
• Disk cache: 24 hour TTL for persistence
• Raw JSONL: Complete API responses saved for debugging

Full-Text Search

The SQLite database includes full-text search indexes on:

• Study titles
• Summaries
• Detailed descriptions

Data Export

Export in multiple formats:

• CSV: Ready for Excel/Google Sheets with key columns
• JSON: Full nested structure preserved
• JSONL: One study per line for streaming/processing

Architecture

• Core API Client: Handles ClinicalTrials.gov API v2 communication
• Database Layer: SQLite with normalized schema and full-text search
• Caching: Two-tier caching (memory + disk) for performance
• MCP Server: Model Context Protocol interface for AI assistants

License

MIT