skills/openai/codex/imagegen/Gen Agent Trust Hub

imagegen

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • Standard Dependency Usage: The skill instructions and scripts refer to well-known packages like openai and pillow. These are standard libraries for interacting with image generation APIs and performing local image processing, and their use aligns with the skill's stated purpose.
  • Secure Credential Management: The skill handles authentication for its CLI mode by utilizing environment variables (e.g., OPENAI_API_KEY). It provides clear documentation on how to set these variables locally rather than prompting for secrets in the chat or using hardcoded credentials.
  • Controlled File Operations: The included Python scripts, such as image_gen.py and remove_chroma_key.py, perform targeted filesystem writes to specific output and temporary directories (e.g., output/imagegen/ and tmp/imagegen/). This behavior is consistent with the need to save generated visual assets for project use.
  • Instructional Integrity: The prompting guidance provided in the reference files focuses on technical optimization and user-intent alignment. There are no patterns suggesting attempts to bypass safety filters or exfiltrate sensitive data beyond the necessary API interaction for image generation.
  • Local Post-Processing: The skill includes a dedicated helper script for background removal, allowing for transparency processing to occur locally using standard image manipulation techniques, which reduces reliance on complex remote fallbacks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 06:59 PM