Analyze Garmin data for meal planning and health patterns
The poster exports Garmin data into a Claude project and meal planner, where Claude spotted a link between eating more on one day and reduced medication efficacy.
Dan Malone runs OpenClaw as an always-on Claude agent connected to Home Assistant through ha-mcp, controlling his lights and fixing his automations.
Dan Malone runs OpenClaw as an always-on Claude agent connected to Home Assistant through ha-mcp, controlling his lights and fixing his automations.
Dan runs OpenClaw in its own LXC container on his Proxmox server, next to Home Assistant, and connects it through ha-mcp. He chats with it over Telegram. When his landing motion lights failed, the agent found two dead Hue bulbs. It couldn't edit the blueprint automation, so it deleted it and rebuilt it as a native one. Dan then asked for weekend nap schedules.
The fix took about five minutes of typing in Telegram. Dan estimates the Home Assistant UI would have taken at least an hour. The agent also updated his dashboard to show the new automation.
Costs add up on per-token API billing, because debugging automations burns a lot of context. Dan rides on his roughly $200/month Claude Max plan, and he runs the agent sandboxed in an isolated container.
Connect to my Home Assistant at [address] and review my automations. Find any that are broken, failing or never triggering, explain what is wrong, and propose fixes. Show me each proposed change and wait for my approval before saving it, then test the ones that are fixed and report the results.
The poster exports Garmin data into a Claude project and meal planner, where Claude spotted a link between eating more on one day and reduced medication efficacy.
A Claude Cowork skill tracks diet and medical metrics in a spreadsheet, and meals told to Claude on mobile are logged in the daily record.
The poster uses Claude with Apple Health, nutrition and lifting app data, plus medication and bloodwork files, and has it maintain an HTML dashboard.