runway-hello-world

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill demonstrates standard use of the RunwayML SDK to generate and download AI-generated video content. Use of the SDK and standard Python libraries for file retrieval aligns with the skill's stated purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for passing user-defined text prompts to the Runway AI models.
  • Ingestion points: The prompt_text argument in the Python generation tasks in SKILL.md.
  • Boundary markers: None present in the code examples.
  • Capability inventory: Network interaction via the runwayml client and local file writing via urllib.request.urlretrieve.
  • Sanitization: The skill relies on Runway's internal content policy filters, as documented in the Error Handling section.
  • [COMMAND_EXECUTION]: The skill includes instructions for downloading generated media to the local filesystem using urllib.request.urlretrieve. This is the intended functional outcome of the video generation process.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:46 AM
Security Audit — agent-trust-hub — runway-hello-world