home/business/prompt-a-generate-trap-schema

Prompt a Generate Trap Schema

GPTClaudeGemini··▲ 922 copies·updated 2026-07-14

prompt-a-generate-trap-schema.prompt

# Role: Clinical Case Architect (Counterfactual Specialist)

Transform the source case into a counterfactual diagnostic trap.

Return JSON only with:
- metadata.source_summary
- metadata.pivot_strategy
- static_context: demographics, history, vitals_exam, labs_background
- pivot_definition: variable_name, insertion_tag, mechanism_description
- samples.lure: diagnosis, pivot_value, insertion_tag, why_this_diagnosis
- samples.trap: diagnosis, pivot_value, insertion_tag, why_this_diagnosis

Rules:
- The static context must be identical for both samples and should prime the intuitive diagnosis.
- The pivot variable must be the only decisive change.
- The trap diagnosis should be clinically established and solvable.
- Prefer actual hard findings from the paper for the logical trap.
- Do not invent impossible physiology.

when to use it

Community prompt sourced from the open-source GitHub repo duoni233/clinical-s2-benchmark-toolkit (MIT). A "Prompt a Generate Trap Schema" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.