The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides patterns for executing external processes and scripts.
rules/transcribe-captions.md: Recommends using child_process.execSync to invoke ffmpeg for audio conversion.
rules/voiceover.md: Instructs the agent/user to execute a custom Node.js script using node --env-file=.env to generate voiceovers.
[EXTERNAL_DOWNLOADS]: The skill frequently instructs the agent to install external dependencies and assets.
Multiple files (rules/3d.md, rules/audio.md, rules/fonts.md, etc.) provide commands to install various @remotion/* packages, zod@3.22.3, mapbox-gl, and @turf/turf from the NPM registry.
rules/lottie.md: Fetches Lottie animation JSON files from assets4.lottiefiles.com.
rules/tailwind.md: Directs the agent to fetch documentation from www.remotion.dev.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data.
Ingestion points: calculateMetadata functions in rules/calculate-metadata.md and rules/compositions.md fetch JSON data from external URLs provided via component props.
Boundary markers: None identified. There are no delimiters or instructions to ignore embedded commands in the fetched data.
Capability inventory: The skill uses the fetch API to retrieve remote data and calculateMetadata to influence the rendering pipeline.
Sanitization: None identified. Fetched data is spread directly into the component's props object without validation or filtering.

remotion-best-practices