skills/runware/runware-skills/ugc-ad/Gen Agent Trust Hub

ugc-ad

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates using vendor-specific tools (runware-run, runware-models) to perform media generation within a secure environment.
  • [EXTERNAL_DOWNLOADS]: Assets referenced in the examples are hosted on the author's official domain (im.runware.ai), representing standard and safe resource usage.
  • [PROMPT_INJECTION]: The skill instructions proactively address security by including explicit guidelines against the creation of deceptive content and ensuring user consent for synthetic likenesses.
  • [SAFE]: Assessment of indirect prompt injection surfaces shows that the skill ingests user scripts and audio assets (ingestion points) for use with the runware-run tool (capability inventory); safety boundaries are established through instructions to verify user-supplied content and adhere to permission policies.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 12:49 PM
Security Audit — agent-trust-hub — ugc-ad