ai-avatar-video

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool restricted to the belt CLI for generating AI video content.
  • [EXTERNAL_DOWNLOADS]: It directs users to install the belt-sh/cli package and references setup scripts hosted on GitHub repositories associated with inference.sh.
  • [PROMPT_INJECTION]: An indirect prompt injection surface exists where user-supplied scripts and URLs are interpolated into commands. • Ingestion points: Fields such as voice_script, prompt, and various URL parameters in SKILL.md. • Boundary markers: No delimiters or warnings are used to isolate user input in the examples provided. • Capability inventory: The skill can invoke the belt CLI tool to perform network-based AI inference tasks. • Sanitization: No methods for escaping or validating external input are described in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 10:40 PM
Security Audit — agent-trust-hub — ai-avatar-video