The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of subprocess.run to execute ffmpeg and ffprobe for video and audio processing. It also uses this mechanism to orchestrate its own internal utility scripts. These operations are essential to the skill's primary function and do not exhibit malicious intent.- [EXTERNAL_DOWNLOADS]: Generation scripts and proxies (single_gen.py, media_proxy.py) download generated video assets and reference media from remote servers. These network operations are part of the standard workflow for interfacing with AI generation services.- [INDIRECT_PROMPT_INJECTION]: The skill processes external JSON configuration files containing brand facts and dialogue. This content is interpolated into prompts for video generation models. While this is an ingestion point for untrusted data, the risk is mitigated by the skill's specific use case, and the generated content is directed at a video model rather than influencing the agent's logic.

render-street-interview