image-review

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes Bash to execute local scripts imgfacts.py and refute.py for image metadata extraction and cross-engine verification. These tools are internal to the vendor environment and align with the skill's purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (image briefs and image content) to perform its review. • Ingestion points: Data enters via the image brief and the visual/metadata content of the images. • Boundary markers: No explicit delimiter or instruction-ignoring syntax is defined for input processing. • Capability inventory: The agent can utilize Bash, Read, Grep, and Glob tools. • Sanitization: There is no documented validation or sanitization of the input brief content.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 06:00 PM