The Agent Skills Directory

[SAFE]: No security issues were detected. The skill provides standard development patterns for the Remotion framework.
[PROMPT_INJECTION]: The skill defines an attack surface for indirect prompt injection as it ingests untrusted data via inputProps to generate video content. This is a characteristic of media generation tools.
Ingestion points: Data enters via the inputProps parameter in rendering functions and through the calculateMetadata hook (rules/calculate-metadata.md).
Boundary markers: Absent. The skill does not instruct on the use of delimiters for user-provided strings used in compositions.
Capability inventory: The skill demonstrates fetching remote data (rules/calculate-metadata.md, rules/lottie.md) and writing files to the local system during rendering.
Sanitization: Absent. Input data is directly interpolated into React components for rendering visuals.
[COMMAND_EXECUTION]: The toolkit describes the use of standard npx remotion CLI commands for rendering videos and deploying serverless functions. These patterns are consistent with the official Remotion documentation and intended use cases.
[EXTERNAL_DOWNLOADS]: The skill correctly references official Remotion packages and reputable services for assets, such as Google Fonts and LottieFiles. All external references are appropriate for the toolkit's purpose.

remotion-video-toolkit