Skills
skills/ace-step/ace-step-1.5/acestep-simplemv/Gen Agent Trust Hub

acestep-simplemv

Fail

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/render-mv.sh acquires administrative privileges using sudo apt-get install to automatically install font packages (fonts-noto-cjk).
  • [COMMAND_EXECUTION]: The script scripts/render.mjs uses execSync to run ffprobe and npx remotion render commands, which involves interpolating variables into shell strings.
  • [REMOTE_CODE_EXECUTION]: The skill setup and rendering workflow uses npm install and npx, which download and execute code from public registries.
  • [EXTERNAL_DOWNLOADS]: The rendering process may automatically download a Chromium-based browser (chrome-headless-shell) from Google servers during execution if no suitable browser is found.
  • [PROMPT_INJECTION]: The skill ingests untrusted user metadata into a video rendering pipeline without robust separation. Ingestion points: scripts/render.mjs via CLI arguments. Boundary markers: Absent. Capability inventory: Command execution via execSync in scripts/render.mjs and system package management in scripts/render-mv.sh. Sanitization: Basic truncation and character replacement for titles.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 17, 2026, 02:55 AM