Oriel — AI & technology worth noticing, every day.

Analyze months of health data via a local MCP server

Web page · https://agentusecases.ai/use-case/analyze-months-of-health-data-via-a-local-mcp-server

Done with Claude. An author connected a local MCP server with 14 tools for blood tests, sleep, CGM and supplements to Claude Desktop, which analyzed six months of biomarkers after a back injury.

First seen by Oriel on .

AI summary

This use case shows Claude Desktop analyzing six months of health data through a locally built TypeScript MCP server with 14 tools, identifying biomarker and sleep trends after a back injury.

  • Amandeep Khurana built the server in about two hours, wrapping Wellavy REST APIs and connecting blood tests, sleep, CGM, and supplement data.
  • Claude analyzed the data in 3–5 minutes, flagged low ferritin and fragmented deep sleep, ranked risk factors, and suggested recommendations.
  • Additional Oura, Whoop, and Libre MCP servers were connected to broaden the available health data.
  • A remote OAuth server failed silently, so API keys in the URL were used instead, routing data through Railway.
  • The suggested prompt requests trend analysis, correlations, plain-language results, a biomarker table, and doctor questions without providing a diagnosis.

#mcp #claude desktop #health data #typescript

Generated by gpt-5.6-sol. Check the original for details.

Aliases

SourceExternal IDURL
Agent Use Caseshttps://agentusecases.ai/use-case/analyze-months-of-health-data-via-a-local-mcp-serverhttps://agentusecases.ai/use-case/analyze-months-of-health-data-via-a-local-mcp-server Read discussion