What problem does it solve? Validating topological data analysis results against published benchmarks and statistical criteria is manual and error-prone; this Skill runs the validation pipeline for a chosen research domain and summarizes pass/fail outcomes with diagnostic context. ## Core Features & Use Cases - Domain-Scoped Validation: Runs validation for financial_tda, poverty_tda, or trajectory_tda via pytest markers and domain-specific scripts. - Benchmark Comparison: Checks results against known references such as Gidea-Katz 2017, 2008 GFC H1 spikes, ARI thresholds above 0.3, and permutation null p-values. - Failure Diagnosis: Reads failing tests to identify whether issues stem from data, topology computation, or statistical tests, then suggests targeted fixes. - Use Case: A researcher suspects the poverty_tda persistence diagrams no longer match known deprivation deciles; the Skill runs the validation suite, reports ARI agreement, and flags suspicious near-zero persistence thresholds. ## Quick Start Ask the assistant to validate the financial_tda domain and summarize which topological benchmarks passed or failed.
