im2-clean-image

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a prompt engineering framework for image generation tools. It functions by providing the agent with instructions and templates to refine user inputs into more effective image generation prompts focusing on material realism and artifact reduction.
  • [SAFE]: No indicators of malicious behavior were found across all categories. Specifically, there are no network requests, file system interactions, credential handling, or code execution capabilities. The instructions are purely text-based guidelines for prompt construction.
  • [SAFE]: The inclusion of pinyin transliterations in the description is legitimate technical terminology related to the skill's primary purpose and does not represent a homoglyph or obfuscation attack.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 01:13 PM