vid-motion-graphics
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to execute a shell script
scripts/export-video.shto process the animation. This script installs Node.js dependencies, downloads Chromium via Playwright, and invokes FFmpeg for video encoding. - [EXTERNAL_DOWNLOADS]: The export script performs automated downloads of the
playwrightNPM package and the Chromium browser engine. These are well-known tools required for the skill's functionality. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it interpolates a user-provided
content_briefdirectly into an HTML template that is subsequently rendered by a headless browser. - Ingestion points: The
content_briefvariable is ingested inSKILL.md(Step 1 and Step 3). - Boundary markers: No specific boundary markers or 'ignore' instructions are provided to separate user content from the HTML/JavaScript logic.
- Capability inventory: The skill can execute shell commands via
export-video.sh, write files to the local disk, and access the network via Playwright to fetch Google Fonts. - Sanitization: There is no evidence of sanitization or escaping of the
content_briefbefore it is placed into the HTML structure, which could allow a malicious brief to inject arbitrary JavaScript into the rendering process.
Audit Metadata