ai-avatar-video

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a documentation resource for the inference.sh platform. It demonstrates the use of the vendor-provided belt CLI tool for media generation tasks without any malicious patterns detected.
  • [COMMAND_EXECUTION]: The skill uses the belt CLI tool within a restricted Bash environment, which is standard for its intended purpose of managing cloud-based AI application runs.
  • [INDIRECT_PROMPT_INJECTION]: The skill enables the processing of external content such as media URLs and text scripts.
  • Ingestion points: Media and text inputs passed to the belt app run command as shown in the examples in SKILL.md.
  • Boundary markers: Inputs are structured within JSON objects, providing a clear separation between the tool's execution instructions and the user-provided data.
  • Capability inventory: The skill utilizes the belt tool to interface with cloud-based AI models hosted by the vendor.
  • Sanitization: The skill instructions do not specify local sanitization, as the content is primarily processed by the vendor's backend API infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 10:24 AM
Security Audit — agent-trust-hub — ai-avatar-video