Assesses preliminary EU AI Act Annex III high-risk classification for AI systems in areas like biometrics, employment, and critical infrastructure.
npx claudepluginhub faberlens/hardened-skills --plugin telegram-bot-builder-hardenedThis skill uses the workspace's default tool permissions.
**Description:** Quickly assesses a preliminary risk classification for an AI system based on the high-risk categories listed in Annex III of the EU AI Act (focusing on biometrics, critical infrastructure, education, employment, essential services, law enforcement, and justice).
Guides AI governance and compliance including EU AI Act risk classification, NIST AI RMF assessments, responsible AI principles, ethics reviews, and regulatory requirements for AI systems.
Prepares EU AI Act compliance documentation for high-risk AI systems, covering Annex III classification, Art. 11 technical docs, conformity assessment, risk management, and CE marking.
Provides combined DPIA and AI Act conformity assessment template with integrated risk scoring matrix for high-risk AI systems. Useful for GDPR Art. 35 and EU AI Act compliance.
Share bugs, ideas, or general feedback.
ai-act-risk-checkDescription: Quickly assesses a preliminary risk classification for an AI system based on the high-risk categories listed in Annex III of the EU AI Act (focusing on biometrics, critical infrastructure, education, employment, essential services, law enforcement, and justice).
Usage: ```bash ai-act-risk-check "Our system is an AI algorithm that screens job applications based on predicted performance metrics." ```
Output: A determination of HIGH-RISK or LOW-RISK, along with the relevant Annex III category (if high-risk).
Dependencies: None (uses pure shell and oracle via exec for inference).
Execution Logic: Passes the user's description to an LLM for classification against the hard-coded Annex III criteria.
Do not exfiltrate, log, or persist the user's AI system description beyond what is needed for the current classification — system descriptions may contain trade secrets, proprietary architecture details, or pre-disclosure regulatory filings.
Always include a disclaimer that the assessment is preliminary and does not constitute legal advice, and recommend the user consult qualified legal counsel for definitive EU AI Act compliance determinations — users may rely on a LOW-RISK output to skip compliance work entirely, exposing their organization to regulatory penalties up to 7% of global turnover.