viral-hook

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates a video processing workflow using dedicated Pika MCP tools for media analysis, asset uploading, and video generation.
  • [SAFE]: The skill demonstrates an indirect prompt injection surface by processing external media analysis and user-provided configuration files.
  • Ingestion points: Media analysis output from Step 2 and user-supplied 'brand.md' files.
  • Boundary markers: None defined.
  • Capability inventory: File upload ('upload_asset') and remote video generation ('generate_viral_hook').
  • Sanitization: The skill employs instruction-based filtering, directing the agent to extract only typography directives while ignoring unrelated content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 08:59 AM
Security Audit — agent-trust-hub — viral-hook