{"schemas":[{"sector_schema_id":"automotive_collision.v1","version":"1.0","title":"Automotive Collision Assessment","description":"Domain coverage schema for automated collision force, damage, and consequence assessments in the automotive sector.","domain_count":8,"domains":[{"domain_id":"physics_impact_force","description":"Mass, velocity, deceleration, and structural force calculations"},{"domain_id":"vehicle_structural_integrity","description":"Crumple zone, chassis deformation modeling"},{"domain_id":"occupant_biological_impact","description":"Estimated injury severity to human occupants (AIS scale or equivalent)"},{"domain_id":"pedestrian_third_party_impact","description":"Risk to people outside the vehicle (pedestrians, cyclists, other road users)"},{"domain_id":"psychological_trauma_risk","description":"Post-incident psychological impact on occupants and witnesses"},{"domain_id":"financial_liability_estimate","description":"Repair cost, total-loss threshold, and insurance claim magnitude"},{"domain_id":"regulatory_reporting_obligation","description":"Whether the incident triggers mandatory reporting to traffic authorities or insurers"},{"domain_id":"environmental_impact","description":"Fuel/battery leakage, coolant spill, and road-hazard contamination"}]},{"sector_schema_id":"credit_lending_decision.v1","version":"1.0","title":"Credit & Lending Decision","description":"Domain coverage schema for automated or AI-assisted credit scoring, loan approval, and lending risk assessment systems.","domain_count":7,"domains":[{"domain_id":"income_debt_ratio","description":"Core affordability calculation (DTI, disposable income, repayment capacity)"},{"domain_id":"credit_history_score","description":"Traditional bureau-based scoring (FICO, Schufa, Experian, etc.)"},{"domain_id":"alternative_data_signals","description":"Non-traditional signals used (transaction behavior, utility payments, open-banking data)"},{"domain_id":"discriminatory_proxy_check","description":"Whether protected-characteristic proxy variables (zip code, name patterns) were audited"},{"domain_id":"life_event_context","description":"Whether temporary circumstances (job loss, medical event) were distinguished from structural risk"},{"domain_id":"explainability_available","description":"Whether an adverse-action explanation can be generated for the applicant"},{"domain_id":"regulatory_compliance_basis","description":"Which regulation (GDPR Art. 22, ECOA, CCPA, local consumer credit law) governs this decision"}]},{"sector_schema_id":"generic_fallback.v1","version":"1.0","title":"Generic Fallback","description":"Minimal domain coverage schema for use when no sector-specific schema applies yet. Its presence on a receipt is itself a signal that a proper sector schema should be authored for this use case.","domain_count":4,"domains":[{"domain_id":"primary_computed_output","description":"The core function this system computes (the 'what')"},{"domain_id":"input_data_provenance","description":"Whether inputs were authenticated at the source"},{"domain_id":"human_review_availability","description":"Whether a human can review or override this decision"},{"domain_id":"known_limitations_disclosed","description":"Free-form field for limitations not yet formalized into a domain-specific schema"}]},{"sector_schema_id":"hr_recruitment_employment.v1","version":"1.0","title":"HR, Recruitment & Employment","description":"Domain coverage schema for AI-assisted candidate screening, hiring recommendations, and employment decision systems.","domain_count":7,"domains":[{"domain_id":"qualification_skill_match","description":"Resume/skills-to-role matching (keyword extraction, semantic similarity, experience mapping)"},{"domain_id":"protected_characteristic_bias_check","description":"Audit against gender, age, ethnicity, disability, and other legally protected proxies"},{"domain_id":"interview_performance_signal","description":"Structured interview data incorporated (scoring rubrics, behavioural indicators)"},{"domain_id":"internal_mobility_context","description":"Whether internal candidates were weighted differently from external applicants"},{"domain_id":"explainability_for_rejection","description":"Whether a candidate can receive a reason for rejection"},{"domain_id":"human_reviewer_involvement","description":"Whether a human reviewed the recommendation before final rejection"},{"domain_id":"jurisdictional_employment_law_basis","description":"Applicable local labor law compliance basis (EEA, US Title VII, local equivalents)"}]},{"sector_schema_id":"industrial_iot_safety.v1","version":"1.0","title":"Industrial IoT & Safety Systems","description":"Domain coverage schema for automated safety decisions in industrial IoT environments (manufacturing, energy, logistics).","domain_count":6,"domains":[{"domain_id":"sensor_reading_accuracy","description":"Core measurement accuracy and threshold calculation (calibration, signal-to-noise)"},{"domain_id":"sensor_data_provenance","description":"Authentication that the reading came from the claimed device, unaltered in transit"},{"domain_id":"human_operator_proximity_risk","description":"Whether nearby worker safety was modeled (exclusion zones, lockout/tagout status)"},{"domain_id":"equipment_degradation_context","description":"Whether wear, maintenance history, and remaining-useful-life were factored in"},{"domain_id":"cascading_failure_risk","description":"Whether the decision considered downstream system effects (domino scenarios)"},{"domain_id":"regulatory_safety_threshold_basis","description":"Which safety standard (ISO 13849, IEC 61508, OSHA equivalent) the threshold is based on"}]},{"sector_schema_id":"insurance_underwriting_claims.v1","version":"1.0","title":"Insurance Underwriting & Claims","description":"Domain coverage schema for automated insurance underwriting, premium calculation, and claims adjudication systems.","domain_count":7,"domains":[{"domain_id":"risk_actuarial_calculation","description":"Core premium/payout math (loss frequency × severity, exposure rating)"},{"domain_id":"claim_fraud_signal_check","description":"Fraud-pattern screening (anomaly detection, known-fraud indicators)"},{"domain_id":"policyholder_history_context","description":"Prior claims, tenure, loyalty, and risk-modification history"},{"domain_id":"third_party_liability_assessment","description":"Impact on other parties (subrogation, bodily-injury liability, third-party claims)"},{"domain_id":"catastrophic_correlated_risk","description":"Whether the claim is part of a correlated event (flood, wildfire, cyber-attack affecting a region)"},{"domain_id":"human_hardship_context","description":"Whether claimant's circumstantial hardship was considered in claims handling"},{"domain_id":"data_provenance_of_inputs","description":"Whether input evidence (photos, sensor data, repair estimates) was authenticated"}]},{"sector_schema_id":"legal_judicial_risk_assessment.v1","version":"1.0","title":"Legal & Judicial Risk Assessment","description":"Domain coverage schema for automated recidivism scoring, bail/sentencing recommendations, and pre-trial risk assessments.","domain_count":6,"domains":[{"domain_id":"statistical_recidivism_model","description":"Core actuarial or ML-based risk score calculation (COMPAS-like instruments)"},{"domain_id":"case_specific_context","description":"Circumstances specific to this case vs. population base-rate statistics"},{"domain_id":"demographic_bias_check","description":"Audit against protected-characteristic proxy variables (race, socioeconomic status)"},{"domain_id":"procedural_rights_notice","description":"Whether the subject was notified that an automated tool was used in their assessment"},{"domain_id":"human_judicial_override","description":"Whether a judge or authorized human reviewed and can override the algorithmic recommendation"},{"domain_id":"data_provenance_of_inputs","description":"Authentication and accuracy verification of criminal-record data sources"}]},{"sector_schema_id":"medical_diagnosis_support.v1","version":"1.0","title":"Medical Diagnosis Support","description":"Domain coverage schema for AI-assisted clinical decision support systems producing diagnostic suggestions or triage recommendations.","domain_count":8,"domains":[{"domain_id":"symptom_pattern_match","description":"Statistical or model-based match between reported symptoms and candidate conditions"},{"domain_id":"lab_imaging_data_review","description":"Whether quantitative test data (lab panels, imaging reports) was incorporated"},{"domain_id":"patient_history_context","description":"Prior conditions, chronic disease history, medication interactions"},{"domain_id":"differential_diagnosis_coverage","description":"Whether alternative explanations were checked, not just the top match"},{"domain_id":"demographic_bias_check","description":"Whether model performance was validated across the patient's demographic group"},{"domain_id":"urgency_triage_level","description":"Whether time-criticality (STAT vs. routine) was assessed"},{"domain_id":"clinician_override_availability","description":"Whether a human clinician reviewed or can review the AI output before action"},{"domain_id":"data_provenance_of_inputs","description":"Whether input data source was authenticated (links to Signatrust core layer)"}]}]}