Skip to main content

MCP AI Services

MCP (Model Context Protocol) AI Services are intelligent engine components in the TIDAS system that enhance lifecycle data understanding, completion, and validation capabilities, designed for next-generation environmental data management and intelligent application scenarios.

✨ Core Features​

MCP services integrate AI capabilities into TIDAS lifecycle databases, providing the following core features:

  • Semantic Search: Query lifecycle inventory (LCA Flows) data using natural language
  • Intelligent Completion: Assist in completing missing fields or suggesting structural improvements
  • Format Validation: Automatically validate data compliance based on JSON Schema
  • Debugging & Validation Support: Supports debugging and validating service processes using the MCP Inspector tool (as shown below)

📷 Illustration: We use the MCP Inspector tool to debug queries for "Cypermethrin" related flow information, verifying that the MCP service successfully returns structured data.

MCP-inspector.pngMCP-inspector.png

🧱 Technical Implementation​

MCP services are implemented based on:

  • Transport Protocol: Supports SSE service with WebSocket, HTTP and other transmission modes
  • Data Processing: Parses and processes data based on TIDAS format (JSON Schema)
  • Tool Scheduling: Integrates various intelligent tools like Search_flows_Tool
  • Local Deployment: Supports private deployment to ensure data security and compliance

🔗 Open Source Link​

MCP services are open-sourced on GitHub:
👉 https://github.com/linancn/tiangong-lca-mcp

The project provides complete server code, sample frontend (including integration examples for the Inspector debugging tool), tool registration and testing documentation, enabling rapid development of AI-driven lifecycle data platforms based on TIDAS architecture.

📦 Prerequisites​

  • A lifecycle database compliant with TIDAS Schema has been built
  • MCP Bearer Token and service address have been configured
  • Required intelligent tools (e.g. Search_flows_Tool) have been registered as needed

If you wish to further integrate LCA intelligent services or build your own AI data platform, MCP services provide an efficient starting point.