pathology-koan-generator

Diagnostic koans are training prompts designed to probe reasoning limits and edge cases. This Skill generates koans that create controlled scenarios testing diagnostic inference, ambiguity handling, and misdiagnosis patterns—useful for pathology education, AI model evaluation, and QA data pipelines.

Edge-case koans target boundary scenarios where diagnostic reasoning typically fails or diverges. Use them in assessment workflows to expose assumption weaknesses, validate robustness under ambiguous conditions, and identify failure modes before deployment.

Yes. Koans serve as learning aids and assessment tools in pathology education. They present structured diagnostic challenges that reveal reasoning gaps, helping learners and systems practice handling uncertainty, conflicting evidence, and non-obvious conclusions.

Koans provide deterministic, controlled test cases for evaluating model reasoning consistency and boundary handling. The Skill generates koan sequences that probe inference quality, enabling systematic QA validation before diagnostic systems reach production.

Diagnostic koans are adversarial learning prompts designed to stress-test assumptions and uncover reasoning fragility. Unlike standard training examples, they deliberately target edge cases, ambiguity, and misdiagnosis scenarios to strengthen robust inference.

The Skill generates koans through an initialized operational context and defined protocol sequence, producing domain-applicable koans without requiring manual authorship. Context setup determines the diagnostic scope; the generation process handles structured koan production.