This is a coding experiment with multi-agent systems, using sport training as the problem domain. It does not replace a coach or sports-medical advice. Use only with a solid training background, at your own risk. No warranty, no support, no audit.
Command
Command "analyse" from airbone42/360-data-athlete, covering /analyse — activity analysis, arguments, workflow, step 1: load athlete knowledge + activity (parallel) and step 2: fetch athlete context.
Command
Scans the coach system for contradictions between config/ files, sub-agents, prompts, the exercise mapping, and external sources (intervals.icu NOTEs). Findings are refined by the config-auditor (fresh context) and written as a markdown report to data/audits/. Fixes go through the config-fixer (fresh context) after…
Command
Shows the current fatigue status of all sport-relevant muscles, grouped by body part.
Command
Fetches and fast-forward-merges the configured default branch from origin into the current working copy. Works against whatever remote is configured (Gitea, GitHub, …) — the command itself is generic.
Command
Resolves a concrete sport-science question with verifiable evidence and persists the finding to framework/research/. Two entry points.
Command
Build the training plan for today (or a given date).
Command
Show current training status without generating a plan.