The Agent Skills Directory

[SAFE]: No malicious patterns detected across all evaluated categories.
[EXTERNAL_DOWNLOADS]: The skill references well-known and trusted external resources for its runtime and test environment.
Fetches animation libraries (GSAP, Lottie) from Cloudflare's CDN and UNPKG.
Loads fonts from Google's official font service.
All external resources are standard for web-based video compositions and do not pose a security risk in this context.
[COMMAND_EXECUTION]: Includes scripts (scripts/render_diff.sh, scripts/frame_strip.sh, assets/test-corpus/run.sh) that use standard tools like ffmpeg, npm, and python3. These are used appropriately for video processing, dependency management, and testing within the skill's intended developer workflow.
[PROMPT_INJECTION]: Instructions in SKILL.md are purely technical and do not attempt to bypass agent safety or reveal system prompts.
[DATA_EXFILTRATION]: No evidence of unauthorized data access or external transmission of sensitive information. The fetch calls found in the test corpus are explicitly defined as blocker examples (patterns the linter must detect to refuse translation).

remotion-to-hyperframes