generate-eu-ai-act-system-card

To generate an EU AI Act system card, invoke the skill with a service path to read MODULE_MANIFEST.md and BEHAVIORAL_CONTRACTS.md, conduct a structured interview, and output a compliance markdown file documenting AI tool usage, risk classification, and human oversight.

You need to document AI system risk classification for EU compliance before the August 2026 deadline, especially when AI tools contribute to code generation, decision support, or automated processing within your services.

Yes, you can specify a risk level when creating AI governance documentation by invoking the skill with the --risk-level flag set to limited, general, or high to guide the risk classification output in the generated system card.

Documenting human oversight mechanisms requires MODULE_MANIFEST.md and BEHAVIORAL_CONTRACTS.md files in the target service directory, which the skill reads to extract service definitions and behavioral constraints for the compliance system card.

The generated AI compliance system card is saved as docs/compliance/eu-ai-act-system-card-<service>-YYYY-MM-DD.md, providing a dated, per-service compliance document for audits, regulator requests, and internal governance packages.

Yes, this compliance documentation approach works for high-risk AI system classifications by accepting the --risk-level high parameter, ensuring the output system card aligns with regulatory requirements for elevated risk categories under the EU AI Act.