image-to-video

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides comprehensive educational content regarding image-to-video generation, including model comparisons, motion prompting techniques, and duration guidelines.
  • [COMMAND_EXECUTION]: The skill includes several example shell commands using the belt CLI (e.g., belt login, belt app run). These are standard usage examples for the documented service and do not involve unauthorized access or dangerous operations.
  • [EXTERNAL_DOWNLOADS]: The skill references an installation script for the belt CLI from a repository associated with the inference.sh service (https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md). As this is a well-known service providing the core functionality described in the skill, this is considered a safe and expected reference.
  • [PROMPT_INJECTION]: No evidence of prompt injection or attempts to bypass AI safety guidelines was found. The instructional text focuses entirely on creative video generation parameters.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 10:39 PM
Security Audit — agent-trust-hub — image-to-video