security-questionnaire-response-review

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed to process external documents (questionnaires), which inherently introduces an indirect prompt injection surface. However, the risk is negligible as the skill possesses no dangerous capabilities such as network access, file system modifications, or code execution.
  • [SAFE]: The instructions include strong security-oriented guardrails, specifically rule sec-q-7, which requires the redaction of secrets, internal vulnerabilities, and confidential customer data during the response preparation process.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 12:11 AM
Security Audit — agent-trust-hub — security-questionnaire-response-review