regulatory-review

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions in SKILL.md (Output Rule 6 and Checklist Item 9) and the associated evaluation criteria in evals.json explicitly direct the agent to suppress common safety disclaimers such as 'This is not legal advice' or 'consult a lawyer'. This represents an attempt to override default safety guardrails that distinguish AI assistance from professional legal services.
  • [NO_CODE]: The skill consists entirely of markdown instructions, reference documentation, and evaluation data, with no executable code, scripts, or external software dependencies provided.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 06:07 AM
Security Audit — agent-trust-hub — regulatory-review