load-anomaly-guard

nadavyigal's avatarfrom nadavyigal

Detects unsafe spikes in training load and emits SafetyFlags with conservative adjustments.

⭐1stars🔀0forks📁View on GitHub🕐Updated Jan 3, 2026
[Guardrails and Safety]

When & Why to Use This Skill

This Claude skill provides a robust safety guardrail for training load management, specifically designed to detect unsafe volume spikes and high monotony in athlete data. By analyzing historical training patterns and injury signals, it emits conservative 'SafetyFlags' and suggests actionable plan adjustments to prevent overtraining and mitigate injury risks, ensuring that AI-driven fitness recommendations remain within safe physiological limits.

Use Cases

  • Automated Overtraining Prevention: Periodically audit athlete training logs to identify week-over-week volume increases exceeding 20-30%, triggering safety alerts before the next session.
  • Post-Workout Risk Assessment: Immediately evaluate high-intensity or long-duration runs to flag potential overexertion and recommend necessary recovery periods or intensity swaps.
  • Injury-Aware Plan Adjustments: Automatically suggest rest days or reduced-volume substitutions when injury flags are detected, prioritizing user health and long-term performance safety.
nameload-anomaly-guard
descriptionDetects unsafe spikes in training load and emits SafetyFlags with conservative adjustments.
short-descriptionBackground load monitor that flags spikes and proposes protective changes.

When Codex should use it

  • Nightly background check on training data.
  • Immediately after a high-intensity or long run is logged.

Invocation guidance

  1. Provide recent TrainingHistory, planned Plan window, and any injury flags.
  2. Compute week-over-week changes and monotony; flag spikes > deterministic caps.
  3. Suggest adjustments (rest/swaps) and emit SafetyFlag[].

Input schema

See references/input-schema.json.

Output schema

See references/output-schema.json.

Integration points

  • Background job: nightly cron.
  • API: v0/app/api/plan/load-guard (new) returning flags + suggested adjustments.
  • UI: Badge on Plan/Today screens; push/email via v0/lib/email.ts.

Safety & guardrails

  • If spike >20–30% week-over-week, emit load_spike and recommend rest or reduced volume.
  • If injury signals present, bias toward rest-day adjustments.
  • No medical diagnosis; advise professional consult on repeated spikes or pain.

Telemetry

  • Emit ai_skill_invoked, ai_safety_flag_raised, and optionally ai_adjustment_applied when suggestions are auto-applied.