design-discriminating-prediction

SkillMedia

Lets your agent design claude skill experiments whose outcomes tell competing hypotheses apart.

Available today. Use it from your connected AI after setup.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Then ask your AI: use the design-discriminating-prediction skill

About this skill

Construct a prediction that yields different outcomes under competing hypotheses.

What this skill tells your AI

The instructions your AI receives, as published by yogsoth-ai/de-anthropocentric-research-engine in skills/design-discriminating-prediction/SKILL.md and read by ahel’s review.

Purpose

Design an observation or test whose outcomes distinguish among competing hypotheses.

Input contract

required: [hypotheses, current_evidence, target_comparison]
optional: [constraints, measurable_variables, candidate_interventions]
constraints: [predictions must differ on an observable outcome]

Procedure

  1. Identify the unresolved contrast between hypotheses.
  2. Derive each hypothesis's prediction under candidate conditions.
  3. Select the condition maximizing interpretable separation within constraints.
  4. Specify outcome, measurement, and interpretation for each possible result.

Output contract

produces: [discriminating_test, prediction_table, measurement_plan, interpretation_rules]
delta_fields: [hypothesis_updates, decisions, uncertainties, recommended_jumps]

Quality gates

  • Predictions differ before the test is selected.
  • Confounds and ambiguous outcomes are recorded.

Failure and counterexamples

Do not call a test discriminating when all hypotheses predict the same result or when the measurement cannot separate them.

Provenance map

  • resolved: design-discriminating-prediction

Signals

GitHub stars
501
Forks
41
Last commit
Sep 2026
Advanced
Catalog kind
skill
Key
design-discriminating-prediction
Source
github.com/yogsoth-ai/de-anthropocentric-research-engine