The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. All external references are to legitimate, well-known video production platforms and open-source frameworks (e.g., GitHub, Remotion.dev, RunwayML).
[SAFE]: Instructions correctly guide the user to handle sensitive data (API keys) outside of the skill instructions, asking if the user has them rather than attempting to harvest them.
[EXTERNAL_DOWNLOADS]: The skill mentions official installation commands for tools like hyperframes and remotion (e.g., npm install hyperframes), which are standard package manager operations for legitimate software.
[DATA_EXFILTRATION]: No network exfiltration or unauthorized file access patterns were detected. The skill primarily provides prompts and workflow guidance.
[PROMPT_INJECTION]: The skill does not contain any instructions that attempt to override system safety guidelines or bypass agent constraints.

video