Stats
Actions
Tags
From running-coach
Computes adherence delta between actual and prescribed workout: pace deviation %, completion %, HR drift, RPE delta. Returns a structured signal dict. Called automatically after log-workout or fetch-strava-activity.
How this skill is triggered — by the user, by Claude, or both
Slash command
/running-coach:analyze-workoutThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Given an actual workout and its prescribed counterpart, computes a structured adherence signal dict using the threshold rules from DESIGN.md §4. The verdict (`on-target`, `under`, `over`, `aborted`) drives the downstream `adapt-plan` strike rules. This skill is never invoked directly by the user — it is always called programmatically by `log-workout` or `fetch-strava-activity`.
Given an actual workout and its prescribed counterpart, computes a structured adherence signal dict using the threshold rules from DESIGN.md §4. The verdict (on-target, under, over, aborted) drives the downstream adapt-plan strike rules. This skill is never invoked directly by the user — it is always called programmatically by log-workout or fetch-strava-activity.
log-workout captures user input.fetch-strava-activity normalizes a Strava activity.A dict {completion_pct, pace_delta_pct, rpe_delta, hr_drift_bpm, verdict} stored in workouts.json[].analysis and returned in-memory to the caller for immediate use by adapt-plan.
See body.md for full instructions and scripts/analyze.py for the deterministic delta computation.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub josix/claude-running-coach --plugin running-coach