remix-graphic-ad-from-reference

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches typography assets from Google Fonts (fonts.googleapis.com and fonts.gstatic.com). These are well-known technology services and are considered safe for design workflows.
  • [COMMAND_EXECUTION]: Executes a local Python script (scripts/cutout_product.py) for image processing and a Node.js utility from the goose-graphics skill for rendering. These operations are restricted to the local environment and the skill's specific output directories (finals/, working/).
  • [REMOTE_CODE_EXECUTION]: No remote code execution from untrusted sources was detected. External image generation is performed via a proxied API to the FAL service (fal-ai/gpt-image-1/edit-image), which is a routine operation for this skill's stated purpose.
  • [PROMPT_INJECTION]: No malicious prompt injection patterns were found. The skill instructions use specific negative constraints (e.g., 'NEVER invent a colour', 'never improvise') to maintain control over the agent's output and prevent off-brand content generation.
  • [DATA_EXFILTRATION]: No sensitive file access or unauthorized network exfiltration was detected. File operations are scoped to image assets and brand configuration provided as input.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 02:14 PM
Security Audit — agent-trust-hub — remix-graphic-ad-from-reference