runway-core-workflow-b
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides legitimate integration with the RunwayML API for AI video generation tasks. It includes Python examples for common workflows like image-to-video animation and style transfer using the
runwaymlSDK. - [SAFE]: No hardcoded credentials, sensitive file access, or suspicious network exfiltration patterns were detected. The skill correctly uses placeholders for external resources and standard Base64 encoding for local image processing.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted data.
- Ingestion points: User-supplied strings and URLs are accepted via
prompt_text,prompt_image, andprompt_videoparameters in the Python SDK examples. - Boundary markers: No explicit boundary markers or "ignore instructions" warnings are present in the code snippets.
- Capability inventory: The skill uses the
runwaymlSDK to perform network operations (API calls) and local file reads (open('photo.jpg')). - Sanitization: No sanitization or validation of the input strings or URLs is demonstrated in the provided examples.
Audit Metadata