image

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill was evaluated across 10 threat categories, including prompt injection, data exfiltration, and remote code execution. No security vulnerabilities were found.
  • [SAFE]: All external tools and APIs mentioned (such as Google Gemini, Black Forest Labs Flux, Ideogram, and Vercel) are well-known, reputable services. No suspicious or obfuscated URLs were detected.
  • [SAFE]: The skill provides command-line examples for image optimization (e.g., cwebp, mogrify) as informational guides for the user. These do not represent autonomous or dangerous command execution by the agent.
  • [SAFE]: The skill reads local context files for product marketing information, which is a standard pattern for context-aware agents. Since the skill does not possess autonomous high-privilege tools, this does not present a significant indirect prompt injection risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 03:58 PM
Security Audit — agent-trust-hub — image