ads-report

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues or malicious patterns were detected in the skill instructions. The content consists of natural language guidelines for an agent to process data and generate reports safely.
  • [PROMPT_INJECTION]: The skill processes external data ("validated Claude Ads JSON run bundle") and renders it into multiple formats, which is a standard surface for indirect prompt injection. The instructions mitigate this by specifying the use of validated data and deterministic templates.
  • Ingestion points: Processes external JSON run bundles (SKILL.md).
  • Boundary markers: Not explicitly defined in the instructions.
  • Capability inventory: Rendering HTML/PDF and writing to a local directory (SKILL.md).
  • Sanitization: No specific sanitization or escaping instructions are provided.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 11:49 PM
Security Audit — agent-trust-hub — ads-report