dbhq-ukdbhq-uk/garmin-skill/skills/garmin/SKILL.md

garmin

Skill0 starsChanged 49 days agoMITView on GitHub

Use for Garmin health and fitness data - body battery, sleep, VO2 max, training load, heart rate, HRV, stress, activities. Also when the user says "garmin".

3 other projects use a version of this.

Adopted by

Most recently taken up by

eddmann/jeevestheir own version, first seen Oct 8, 2026
ai-ecoverse/skillstheir own version, first seen Oct 2, 2026
davidmosiah/garmin-mcptheir own version, first seen May 4, 2026

Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
davidmosiah/garmin-mcp★ 13same name
eddmann/jeeves★ 2same name
ai-ecoverse/skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repositoryOne other file it gives its agents.
- AGENTS.mdAGENTS.md

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

0
Common
Most of the 3 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
Use Garmin's names and labels as they come ("Body Battery", "Productive").
line 59
Distinctive:
Never diagnose, or suggest a condition, from these numbers.
line 61

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

SKILL.md

Garmin Health & Fitness

Reads the user's Garmin Connect data - vitals, sleep, activities, training status - live, or archived to markdown.

When not to use

Anything that writes to Garmin: logging an activity, editing a workout, changing a setting. Every script only reads.
Medical questions. See "Reading the numbers".

Commands

${CLAUDE_SKILL_DIR}/.venv/bin/python ${CLAUDE_SKILL_DIR}/scripts/garmin_health.py today       # or yesterday, YYYY-MM-DD, week
${CLAUDE_SKILL_DIR}/.venv/bin/python ${CLAUDE_SKILL_DIR}/scripts/garmin_sleep.py              # last night; or yesterday, YYYY-MM-DD
${CLAUDE_SKILL_DIR}/.venv/bin/python ${CLAUDE_SKILL_DIR}/scripts/garmin_activities.py 7       # activities, last N days
${CLAUDE_SKILL_DIR}/.venv/bin/python ${CLAUDE_SKILL_DIR}/scripts/garmin_activities.py training
${CLAUDE_SKILL_DIR}/.venv/bin/python ${CLAUDE_SKILL_DIR}/scripts/garmin_snapshot.py --output-dir DIR 2026-02-22
${CLAUDE_SKILL_DIR}/.venv/bin/python ${CLAUDE_SKILL_DIR}/scripts/garmin_rollup.py --output-dir DIR 2026-W08
${CLAUDE_SKILL_DIR}/.venv/bin/python ${CLAUDE_SKILL_DIR}/scripts/garmin_status.py             # offline: settings, tokens, cooldown

health: resting HR, HRV, Body Battery (the day's lowest to highest level, and the latest), stress, steps, calories. week is a 7-day table, and takes a few seconds.
training: VO2 max, training load, readiness, status.
--json on health, sleep or activities: the same figures, unformatted, with null for no data.
snapshot writes a day to DIR/YYYY-MM-DD.md (default today). rollup writes an ISO week to DIR/YYYY-Www.md (default this week; last works). Days after today are left empty and not fetched.

"No data" in a row means Garmin has nothing for that day. It is not an error.

Settings

~/.dbhq/garmin/config.json holds the account email and one preference, "units": "imperial" (default, miles) or "metric" (km). It changes activity distances only. No password is stored: the login asks for it.

When a script fails

The scripts only resume a saved session. None of them logs in. If the cause is unclear, run garmin_status.py: it never calls Garmin. Match the message:

Message	What to do
`Config file not found`	Ask the user to run `${CLAUDE_SKILL_DIR}/scripts/setup.sh` in their own terminal
`Not authenticated`, `Old token format` or `Could not resume Garmin session`	Ask the user to log in, below
`rate-limiting`, with a time	Wait until that time. Do not log in or retry: every script refuses until then, and another attempt extends the block
`Garmin refused the login`	Wrong password: the user logs in again. Wrong email: the user reruns `setup.sh`
`Error: ... failed`	The call failed. Snapshot and rollup wrote nothing. Say it failed, never "No data"

The login, for the user to run in their own terminal. Do not run it yourself: it needs a terminal, and asks for an MFA code if Garmin sends one.

${CLAUDE_SKILL_DIR}/.venv/bin/python ${CLAUDE_SKILL_DIR}/scripts/garmin_login.py

Reading the numbers

They are estimates from a wrist sensor. Body Battery, stress, sleep stages, training readiness and training status are Garmin's own models, not measurements.
Use Garmin's names and labels as they come ("Body Battery", "Productive"). Do not invent a score or a scale.
Compare against the user's own recent days (garmin_health.py week, earlier snapshots), not population norms.
Never diagnose, or suggest a condition, from these numbers. If the user is worried by a reading, suggest they talk to a clinician.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free