The Agent Skills Directory

[SAFE]: The skill follows established software development and media rendering best practices. No malicious code, obfuscation, or injection attempts were detected.
[DATA_EXPOSURE_AND_EXFILTRATION]: File operations are limited to project-local directories for outputting frames and videos (e.g., output/). There is no evidence of sensitive data harvesting or exfiltration to external domains.
[COMMAND_EXECUTION]: The skill uses standard CLI tools such as tsc, ffmpeg, and ffprobe for legitimate build and inspection tasks. These operations are restricted to the local environment and are expected within a rendering workflow.
[INDIRECT_PROMPT_INJECTION]: The skill defines an interface for ingesting user feedback via notes.json during the preview phase. While this represents a data ingestion surface, it is a documented feature for collaborative review and does not exhibit malicious patterns.
[DYNAMIC_EXECUTION]: The startPreview function dynamically loads a composition module from a local file path. This is a standard architectural pattern for modular media engines and is implemented using project-relative paths.

framefields-render