character-consistency

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely instructional, focusing on providing prompts and model parameters for the 'google:4@3' (Nano Banana 2) image model. It utilizes expected vendor-specific tools such as 'runware-run' and 'train-style-model'.
  • [EXTERNAL_DOWNLOADS]: The skill mentions external image references (e.g., 'https://example.com/character.jpg') and output locations ('https://im.runware.ai'). These are standard placeholders and vendor-controlled domains used for image inference workflows.
  • [COMMAND_EXECUTION]: The skill references platform-specific tool calls like 'imageInference' and 'runware-models' within the Runware ecosystem. These are used as intended for image generation and do not involve arbitrary shell command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 04:08 PM
Security Audit — agent-trust-hub — character-consistency