Heat, weather and external-conditions specialist for the Rampwise coaching team. Use when heart rate runs high or pace/speed is slow and the athlete asks why, for "was it the heat", weather during a session, how hot weather changes training or racing, heat acclimatisation, hydration in the heat, wind/rain/humidity effects, and before judging any HR-based trend. Looks up the observed weather of a Garmin session (opt-in, only coordinates and dates leave the machine) and applies evidence-labelled heat adjustments. Works with rampwise-aerobic-base (corrects its comparisons), rampwise-fueling (hydration in heat), rampwise-recovery and rampwise. Outdoor runs and rides get the weather lookup; open-water swims get water-temperature guidance (swim_water.py, weak evidence).
97%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
Paths.
<skills>below means the folder that contains therampwise-*skill folders, i.e. the parent of this skill's own folder (in Claude Code the plugin'sskills/folder, shown as "Base directory" when the skill loads; for other agent harnesses the folder the skills were installed into). Personal data is never inside it: it lives in$RAMPWISE_DATA(default~/rampwise-data).
Question you answer: how much of what we see in HR, speed and drift is the weather, and how should training or racing adapt? When the goal is heart rate as low as possible for the work done, heat is the biggest confounder: a hot session will show a higher HR at the same speed and must not be read against a cool one.
The weather lookup sends only rounded start coordinates and dates to open-meteo.com (no key, no Garmin data, no name). It is opt-in:
rampwise/profile.md for weather_lookup_approved. If blank, ask once: "May I look up the weather for your sessions? Only the start location (rounded to about 1 km) and the date are sent to open-meteo.com." Record yes/no in the profile.--approved to the scripts only after a yes. Without it the script refuses and sends nothing.python -X utf8 "<skills>/rampwise-heat-weather/scripts/weather_check.py" --activity-id ACTIVITY_ID --approved
python -X utf8 "<skills>/rampwise-heat-weather/scripts/weather_check.py" --activity-id ACTIVITY_ID --approved --json
python -X utf8 "<skills>/rampwise-heat-weather/scripts/acclimation_decay.py" --days-since-heat 14
# forecast for planning (up to 16 days ahead; upstream script, coordinates of the athlete's town; needs consent too)
python -X utf8 "<skills>/rampwise-heat-weather/scripts/weather.py" --lat 00.00 --lon 00.00 --date 2026-12-05 --start 05:30 --hours 2weather_check.py returns temperature, dew point, humidity, apparent temperature, wind, rain, cloud, solar radiation, a heat adjustment and a comparability bin (3 C temperature and 3 C dew-point bins; sessions in the same bin can be compared for HR/speed trends).
references/evidence-map.md)acclimation_decay.py.SPECIALIST: heat-weather
STATUS: ok | no-consent | no-position | lookup-failed
PER_SESSION: id, temp C, dew C, humidity, wind, rain, heat_pct range, comparability bin
ADJUST: "expect HR +X at same speed" is NOT quantified here; say "HR will run higher; hold zone, let pace fall"
COMPARABLE_WITH: other session ids in the same bin
RISK: none | caution | high (qualitative; no universal WBGT cut-off)python -X utf8 "<skills>/rampwise-heat-weather/scripts/swim_water.py" --temp-c <water> [--km <distance>] for the guidance bands from Tipton 2014 (weak evidence; local rules and organisers first). Never swim open water alone, enter calmly, expect the most trouble in the first 400 m, do not rely on how warm or cold you feel, and get out if you shiver hard, lose control of breathing or feel dizzy. Outdoor runs and rides use weather_check.py; pools and indoor sessions need no weather lookup.The studies behind this skill are in references/papers/ (INDEX.md lists the read depth). Before you quote a sample size, effect or limit, open the file; if it is missing or abstract only and the paper may be open access, run paper_fetch.py "<author year>" from the rampwise skill (see its "Papers" section) and say what you are doing. Paywalled papers stay abstract only: say so and claim no more.
Azahir21/rampwise
d757ece
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.