remotion-best-practices
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill instructions facilitate the ingestion and processing of external, untrusted data (JSON) which presents an attack surface for indirect prompt injection.
- Ingestion points: Data is fetched from remote URLs in
rules/lottie.md(Lottiefiles API) andrules/calculate-metadata.md(dynamicprops.dataUrl), or loaded from local files inrules/display-captions.md(captions JSON). - Boundary markers: Explicit delimiters or instructions to ignore embedded commands are absent from the prompt interpolation patterns.
- Capability inventory: The execution environment allows for shell command execution (
bunx,npx), file system operations (fs.writeFileSyncinrules/transcribe-captions.md), and network requests (fetch). - Sanitization: The skill lacks explicit sanitization or schema validation steps for fetched data before it is utilized in rendering or metadata calculation.
- [EXTERNAL_DOWNLOADS]: The skill references and recommends downloading sound effect assets from an unverified third-party GitHub repository (
kapishdima/soundcn) inrules/sfx.mdwithout providing integrity verification mechanisms.
Audit Metadata