trend-format-replication

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input when deconstructing trending formats or viral templates to generate new content. If these inputs contain malicious instructions, they could influence the agent's behavior during the video generation process.
  • Ingestion points: User requests referencing specific trending audio, viral templates, or trending short-form formats (SKILL.md).
  • Boundary markers: The instructions lack explicit delimiters or instructions for the agent to ignore potentially malicious content embedded within the user-supplied trend descriptions.
  • Capability inventory: The skill interacts with video generation models like 'flux-3-video-draft' (SKILL.md).
  • Sanitization: No sanitization or validation logic is defined for the external descriptions ingested by the skill.
  • [EXTERNAL_DOWNLOADS]: The skill's setup section provides instructions to connect to a remote Model Context Protocol (MCP) server.
  • Source: 'https://mcp.apimage.org/mcp'.
  • This resource is hosted on the infrastructure of the skill's author ('apimageorg').
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 12:35 AM