ugc-haul-format

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill configuration process involves adding a Model Context Protocol (MCP) server from https://mcp.apimage.org/mcp. This domain corresponds to the skill vendor and provides the necessary tools for the video generation workflow described in the instructions.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external product assets (images) for video generation, which represents an ingestion surface for third-party data.
  • Ingestion points: Product image assets are passed as references to the generate_video function within the production loop in SKILL.md.
  • Boundary markers: The skill employs a consistent system prompt and fixed parameters (LOOK, SEED) to maintain control over the generated output.
  • Capability inventory: The skill utilizes a generate_video tool call to create video files from the provided assets.
  • Sanitization: No explicit sanitization is performed on asset metadata, though the scope of the operation is restricted to media generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 09:15 AM