remotion-best-practices

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides numerous commands using package managers ("npm", "yarn", "pnpm", "bun") and "npx" to install Remotion-specific packages and run Remotion CLI tools (e.g., "npx remotion studio", "npx remotion render"). These are standard operations for setting up and using a Remotion project.
  • [EXTERNAL_DOWNLOADS]: The skill references several external resources, all of which are well-known services or vendor-controlled domains:
  • Fetches audio and video assets from "remotion.media" (vendor domain).
  • Integrates with "api.elevenlabs.io" for AI voiceover generation.
  • Loads Lottie animations from "assets4.lottiefiles.com" (LottieFiles).
  • Uses Mapbox APIs via "mapbox-gl" for map animations.
  • Downloads Whisper.cpp models using the "@remotion/install-whisper-cpp" package, which is a common utility in the Remotion ecosystem for generating captions.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes patterns for ingesting data from external APIs (e.g., in "calculateMetadata" and "voiceover.md"). While this technically creates a surface for indirect prompt injection if the external data is malicious, the skill demonstrates best practices such as using Zod for schema validation ("parameters.md") to sanitize inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 03:08 PM