eu-ai-act-triage-en

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a robust 'Human Gate' mechanism, requiring all draft triage cards to be reviewed and approved by a qualified professional (e.g., a lawyer or compliance officer) before being used.
  • [SAFE]: The instructions establish a 'Step 0' verification process that mandates fetching current legal text from official sources like EUR-Lex, preventing the agent from relying solely on potentially outdated or incorrect training data for legal citations.
  • [SAFE]: The skill's configuration is privacy-conscious, specifying that data residency is local and PII egress is prohibited.
  • [SAFE]: The integration with the 'mcp-eu-compliance' connector and other companion skills (e.g., 'matematic-konstytucja-ai') is consistent with the stated purpose of building a compliance record-keeping ecosystem within the vendor's own framework.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 08:27 AM
Security Audit — agent-trust-hub — eu-ai-act-triage-en