klingai-compliance-review

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reviewing user-provided artifacts or "canaries" for policy compliance. Processing these external inputs constitutes an indirect prompt injection surface.
  • Ingestion points: Artifacts and review scopes described in the instructions section of SKILL.md.
  • Boundary markers: None explicitly defined in the provided code snippets or instructions.
  • Capability inventory: Access to Read, Write, Edit, Bash, and Grep tools to facilitate the review process and report generation.
  • Sanitization: The instructions contain explicit mandates to reject artifacts containing private information, unlicensed material, or restricted content, acting as a procedural safeguard.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:45 AM
Security Audit — agent-trust-hub — klingai-compliance-review