The Agent Skills Directory

[PROMPT_INJECTION]: The skill documents patterns for fetching and processing untrusted external data, such as Lottie animation files, SRT subtitle files, and dynamic metadata from external APIs. This creates a surface for indirect prompt injection if the source content contains malicious instructions.\n
Ingestion points: rules/lottie.md, rules/import-srt-captions.md, rules/calculate-metadata.md, and rules/compositions.md describe fetching content via fetch() and staticFile().\n
Boundary markers: The provided code examples lack explicit delimiters or instructions to ignore embedded prompts in external data.\n
Capability inventory: Fetched data is used to drive animations, populate component properties, and determine composition metadata (duration, dimensions).\n
Sanitization: No data validation or sanitization logic is present in the provided examples for handling fetched content.\n- [EXTERNAL_DOWNLOADS]: The skill references and retrieves resources from legitimate external services and official documentation sites.\n
Fetches motion assets from LottieFiles, an established service for web animations.\n
Directs users to the official Remotion documentation for Tailwind CSS integration instructions.\n
Uses placeholder URLs for demonstrating asset loading and API integration.

remotion-best-practices