decision-hygiene

Core decision analysis skill. Trigger whenever the user: describes a decision, choice, or dilemma in natural language ("I need to choose between...", "we're torn between...", "should we...", "I'm not sure whether to...", "help me decide"); faces a trade-off between options; asks to triage, structure, or analyse a decision; asks for noise audit, calibration check, or bias catalog lookup; uses phrases like "decision hygiene" or "triage this decision". Trigger proactively — do not wait for technical vocabulary. Also activates for: structured scoring of options, judgment variance measurement, prediction tracking, heuristic pattern queries.

Decision Hygiene Skill

What this skill provides

Reference catalogs for structured lookups (read on demand):

bias-catalog.yaml — cognitive biases with detection questions (questioner agent)
fallacy-catalog.yaml — logical fallacies with structural patterns (logic-tester agent)
heuristic-patterns.yaml — common heuristic activations
inhibition-strategies.yaml — bias inhibition approaches

Inline computation formulas (no scripts required in Cowork):

Triage — stakes/complexity scoring → protocol routing (in sentinel.md)
Noise — divergence, CV, IQR algorithms (in noise-calculator agent)
Calibration — Brier score, overconfidence by bucket (in calibration-coach agent)

Important note on catalogs

The interaction_multipliers in bias-catalog.yaml are design parameters, NOT empirical values. They indicate relative severity, not measured effects.