generate-video

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill provides an indirect prompt injection surface as it accepts user-provided text prompts and interpolates them into instructions sent to external generative AI models.
  • Ingestion points: User-provided prompt captured in scripts/generate.ts.
  • Boundary markers: Absent; user input is directly combined with style hints.
  • Capability inventory: The skill can perform network requests to external APIs, write files to the local file system, and execute internal scripts via bun.
  • Sanitization: Absent.
  • [COMMAND_EXECUTION]: The skill uses the bun runtime to execute its own scripts and helper tools, which is the intended method for providing video generation functionality.
  • [EXTERNAL_DOWNLOADS]: The skill initiates network connections to well-known and reputable services, including the Google Gemini API and Replicate, to process video and image generation requests.
  • [REMOTE_CODE_EXECUTION]: The skill's main script performs dynamic path resolution and imports to locate shared vendor utilities, including reading local configuration from the agent's plugin directory to determine installation paths.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 04:02 AM