make-ai-short

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes several local utility scripts via the command line, including python tools/gen_voice.py, python tools/gen_image.py, and python tools/gen_clip.py to interface with external AI services. It also uses npm run gen for video composition and rendering. These actions are aligned with the skill's stated purpose of automating a video production pipeline.
  • [INDIRECT_PROMPT_INJECTION]: The workflow involves taking user-provided descriptions and scripts to generate prompts for video and image models. While this presents an inherent surface for prompt-based manipulation of the generated media, the skill includes explicit rules for consistency and quality assurance (such as the 'three iron rules' and manual QA gates) which serve as functional mitigations for unintended output.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 07:41 PM