authority-triage-en

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains purely instructional content for legal research and source hierarchy mapping with no executable scripts or system commands.
  • [SAFE]: No remote code execution, external downloads, or persistence mechanisms were detected. External tools mentioned (e.g., eu-sparql-search) are standard Model Context Protocol (MCP) connectors used for data retrieval.
  • [SAFE]: The skill implements safety best practices by defining a 'Human Gate' section, which mandates that the agent's output be treated as a draft requiring qualified human approval before use.
  • [SAFE]: It includes specific instructions to tag unconfirmed citations with '[TO VERIFY]', reducing the risk of the agent presenting hallucinations as factual legal provisions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 08:27 AM
Security Audit — agent-trust-hub — authority-triage-en