The Agent Skills Directory

[PROMPT_INJECTION]: The skill describes patterns for fetching and processing external data (JSON, SRT, Lottie files) that could be leveraged for indirect prompt injection.
Ingestion points: rules/calculate-metadata.md (fetching remote JSON), rules/display-captions.md (loading captions JSON), rules/import-srt-captions.md (parsing SRT subtitles), and rules/lottie.md (fetching animation data).
Boundary markers: The code examples do not provide delimiters or instructions to ignore commands that may be embedded in the retrieved data.
Capability inventory: The skill utilizes subprocess execution for media tasks (bunx remotion ffmpeg in rules/ffmpeg.md) and file system writes for asset creation (fs.writeFileSync in rules/transcribe-captions.md).
Sanitization: No data validation or sanitization logic is shown for the external content before it is processed by the application logic.
[COMMAND_EXECUTION]: rules/ffmpeg.md documents the use of bunx remotion ffmpeg and bunx remotion ffprobe for video trimming and metadata retrieval.
[EXTERNAL_DOWNLOADS]: The skill encourages the use of various external resources and services, including assets and documentation from well-known domains such as remotion.dev, mapbox.com, elevenlabs.io, and lottiefiles.com.

remotion-best-practices