Let wearable data guide workouts and recovery

One user had Claude turn Garmin history and health signals into a daily training plan shown on a phone-friendly webpage.

Done with Claude· Advanced· Oct 2026

What they did
They set up hourly Garmin syncing into an InfluxDB database on their own server, then scheduled Claude Code to run each morning. The agent looked at sleep, stress, load, and similar measures, created its own readiness score, and used that to revise the day’s training. It then published the updated plan on a small webpage they could check from a phone, using a large back catalog for context.
How it went
They said the routine helped keep workouts specific enough to lower the risk of burnout or overtraining. It also seemed to catch lower-energy stretches such as travel, holidays, and long work weeks, but they did not report any quantified outcome.
Tips
A setup like this depends on reliable data collection and an automatic job running on your own server. It also works better if you give it a substantial history up front, not just a few recent days.

Try it yourself with Claude

I’m going to share [wearable history], [training app logs], [nutrition records], and any [bloodwork or medication notes] that may matter. Review the patterns, estimate my current training readiness, and update a simple HTML dashboard I can open on my phone. Using that information, suggest a training plan for [goal] and list three practical changes to try this week. Stop and ask me before you do anything irreversible, including deleting data, sending a message, or submitting a change, and separate anything medical from anything you can safely infer yourself.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Read the original on reddit.com ↗

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.