ad-creative

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates within its stated marketing purpose. The provided Python script scripts/ad_copy_validator.py processes data locally for validation purposes and does not perform network operations or access sensitive system files.- [PROMPT_INJECTION]: The skill includes an iteration mode that processes existing ad copy and performance data, creating a surface for indirect prompt injection. However, the skill lacks external network access or file-writing capabilities that would allow for data exfiltration or persistence.
  • Ingestion points: SKILL.md (Iteration methodology) and standard input/file input for scripts/ad_copy_validator.py.
  • Boundary markers: None present in instructions.
  • Capability inventory: Local script execution and text generation output.
  • Sanitization: Script validates against marketing compliance rules but does not sanitize for instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 01:52 AM
Security Audit — agent-trust-hub — ad-creative