The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses npx and other package runners to execute Remotion-specific CLI tools for project initialization, previewing, and rendering (e.g., create-video, remotion studio, remotion still, remotion ffmpeg).
[EXTERNAL_DOWNLOADS]: The skill recommends installing numerous packages from the Remotion ecosystem and third-party libraries like zod and maplibre-gl. It also provides instructions for using @remotion/install-whisper-cpp to download and install the Whisper C++ binary for audio transcription.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes data from external sources, which could potentially contain malicious instructions that influence the agent's behavior during the video creation process.
Ingestion points: API responses fetched in rules/calculate-metadata.md, JSON caption files in rules/display-captions.md, SRT subtitle files in rules/import-srt-captions.md, and TTS API responses in rules/voiceover.md.
Boundary markers: Absent; the skill does not specify the use of clear delimiters or instructions to ignore content within the ingested data.
Capability inventory: The skill possesses capabilities for command execution (Remotion CLI tools), file writing (transcription output), and network operations (fetching assets and metadata).
Sanitization: The skill recommends using Zod schemas for parameter validation as a best practice in rules/parameters.md.

remotion-best-practices