skills/whytryharder/skills/google-veo/Gen Agent Trust Hub

google-veo

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the installation of the belt CLI and additional skill packages from the belt-sh and inference-sh organizations. These resources are necessary for the skill's primary function of video generation through the inference.sh API.
  • [COMMAND_EXECUTION]: The skill utilizes the belt command-line utility to run video generation models. Execution is strictly scoped to the belt binary via the allowed-tools frontmatter configuration.
  • [PROMPT_INJECTION]: The skill takes user-provided prompts and interpolates them into JSON payloads for shell commands. This represents an indirect prompt injection surface.
  • Ingestion points: User prompts in SKILL.md examples.
  • Boundary markers: Commands use JSON structure within single quotes to delimit input.
  • Capability inventory: Limited to the belt CLI tool via Bash.
  • Sanitization: Relies on the belt CLI's internal handling of JSON inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 05:03 AM
Security Audit — agent-trust-hub — google-veo