io.github.mdfifty50-boop/compliance-shield
EU AI Act compliance audit trails and evidence generation for AI agent systems — risk classification
Versions
0.1.0latestTools 6
assess_risk_level Classify an AI system under the EU AI Act risk framework (Unacceptable, High, Limited, Minimal). Returns required obligations and article references.
create_audit_trail Start a new compliance audit trail for an AI system. Returns a trail_id used for all subsequent logging and compliance checks.
log_decision Log an AI decision with full traceability — input, output, reasoning, human oversight status, model used, and confidence. Each entry is timestamped and auto-numbered.
check_compliance_gaps Identify missing EU AI Act compliance requirements for an audit trail. Checks human oversight, decision logging, reasoning documentation, model identification, and responsible person assignment.
generate_evidence_package Generate an auditor-ready evidence package with system info, decision log summary, human oversight statistics, model usage breakdown, and compliance status. Suitable for regulatory submission.
get_enforcement_timeline Show upcoming EU AI Act enforcement deadlines with days remaining. Includes penalty structure for violations (up to 35M EUR / 7% global turnover).
Permissions 1
env_vars low