seedream-image

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Category 8: Indirect Prompt Injection] (SAFE): The skill processes user-supplied text for image generation prompts.
  • Ingestion points: The prompt and reference_image inputs in SKILL.md.
  • Boundary markers: Absent; however, the impact is limited to the content of generated images.
  • Capability inventory: CLI-based workflow execution, file downloads, and system file opening.
  • Sanitization: Uses the Seedream API's enable_safety_checker parameter.
  • [Category 10: Dynamic Execution] (SAFE): The skill documentation provides shell script examples to download and automatically open resulting files.
  • Evidence: SKILL.md includes a bash loop that uses the system open command on files downloaded to the Desktop.
  • Context: This behavior is consistent with the skill's primary purpose of generating and displaying visual assets.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:25 PM