ai-marketing-videos

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the belt CLI tool to execute video generation, audio synthesis, and media merging tasks on the inference.sh platform.
  • [PROMPT_INJECTION]: Indirect prompt injection surface identified in the Explainer Video workflow. 1. Ingestion point: Script content generated by Claude and saved to script.json. 2. Boundary markers: Absent. 3. Capability inventory: belt app run (network-enabled AI inference) across all workflow steps. 4. Sanitization: Absent; model-generated text is interpolated directly into downstream prompts for Kokoro-TTS and media-merger.
  • [EXTERNAL_DOWNLOADS]: References installation scripts and media assets hosted on inference.sh and github.com/inference-sh, which are identified as the vendor's primary infrastructure for this skill.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 09:23 AM
Security Audit — agent-trust-hub — ai-marketing-videos