I just built a MCP Server that connects Claude to all your wearables
MCP server for wearables, but health data privacy concerns remain.

Normalizes Whoop, Garmin, and Withings data into one place for Claude to analyze.
Biohackers, athletes, data-driven health enthusiasts
Exist.io · Quantified Self tools · Apple Health
A few months ago I posted here to announce Pace. A lot of very helpful technical questions about the JSONB approach, MCP streaming, rate limits. Thanks and I took notes ;)
Since then I have reworked most of the data layer, so better normalization when 2 or more providers are connected, more enriching tools for samples (HR, power, GPS) and more additional tools.
Pace is a Claude MCP connector that lets you analyse your wearable data. Once you connect your at our website, you can ask Claude things like:
- How was my sleep? - Compare my runs this week vs last week? - What's my HRV trend over the last 30 days?
Claude handles the analysis and visualization. I don't touch dashboards anymore.
My colleague for example uses Whoop for sleep, Garmin for cycling and Withings for weight. Everything normalized in one place.
Free to try, no Pro/MAX plan needed.
Happy to answer questions
MCP server for wearables, but health data privacy concerns remain.
MCP wrapper around Oura's existing API when direct API calls work just as well.
Unified wearable API across 5+ providers, but the AI layer remains vaporware and concept is crowded.
Persistent memory for Claude Code with auto-compaction recovery—but relies entirely on external API.
MCP wrapper around AppControl — Windows only, requires the desktop app running.
MCP server lets Claude query your life data—demo mode with 4,251 entries means instant try-before-commit.