model-wearing-product

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides setup instructions in the documentation for the user to manually register an MCP server using the claude mcp add command.
  • [EXTERNAL_DOWNLOADS]: The skill references an external endpoint at https://mcp.apimage.org/mcp which is used during setup to fetch the tool definitions and service logic for the associated image generation tools. This endpoint belongs to the skill's vendor infrastructure.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes a workflow that ingests user-provided text prompts and reference images, which represents a surface for potential injection targeting the image generation tools.
  • Ingestion points: Prompts and reference_images arrays defined in the generate_image examples within SKILL.md.
  • Boundary markers: The instructions do not specify explicit delimiters or "ignore embedded instructions" warnings for the interpolated data.
  • Capability inventory: The skill utilizes image generation, asset creation, and background replacement tools (generate_image, create_brand_asset, replace_background).
  • Sanitization: No explicit sanitization, validation, or filtering of the prompt inputs is documented in the skill body.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 12:35 AM