Has anyone used Claude as a training and diet coach with real wearable data?
A Reddit user is exploring using Claude as a personal training and diet coach, leveraging real wearable data. The user plans to integrate health data from two smartwatches and a phone, all syncing to Health Connect on Android 16. This data would then be piped to a personal server and through a custom MCP connector, allowing Claude to query and provide tailored advice. The user refined this concept with Claude's help, which also assisted in structuring and writing the Reddit post.
时间与来源
时间显示为 UTC
显示时区：UTC
本地时区尚不可用，暂时显示 UTC。
发布当时偏移：UTC+02026年10月10日 10:33 UTC
收录当时偏移：UTC+02026年10月10日 16:00 UTC
- 发布
- 2026年10月10日 10:33
- 收录
- 2026年10月10日 16:00
- 来源类型
- 开发者社区
- 档位
- 社区
- 信源状态
- 正常
档位是按信源手工设定的编辑判断，不是逐条打分。
I want Claude to act as my personal coach: read my health data (sleep, heart rate, steps, workouts) and adjust my training and diet plans based on it.
My setup: two watches (CMF Watch from Nothing for workouts, a Timberland watch for daily wear) plus my phone, all syncing to Health Connect on Android 16. I have a home lab and a VPS, so I’m thinking of building a small pipeline: Health Connect → my own server → a custom MCP connector so Claude can query the data when I ask.
Before I build anything:
- Has anyone connected wearable or Health Connect data to Claude? Which connector or MCP server did you use?
- Does it actually give useful, adaptive advice, or does it get generic after a while?
- How do you handle memory across conversations, so Claude keeps track of the plan and progress over weeks? Projects, a skill, something else?
- Any concerns about sharing health data this way, and how did you deal with them?
Genuine question about my own setup, not promoting anything. Refined this idea in a conversation with Claude and used it to help structure and write the post in English (I’m Portuguese).