The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary function is to generate and execute ffmpeg commands. The instructions provided in SKILL.md and the reference files guide the agent to use specific filter graphs defined in the skill's asset JSON files to process video clips.
[COMMAND_EXECUTION]: The demo helper script examples/demo/run-preset.mjs uses node:child_process spawnSync to invoke ffmpeg. It passes arguments as an array rather than a shell string, which is a secure practice that prevents shell injection via filenames or other parameters.
[EXTERNAL_DOWNLOADS]: The documentation in examples/demo/README.md provides a command to download a Creative Commons licensed movie from a well-known mirror (mirrors.dotsrc.org) for demonstration purposes. This is a legitimate resource used for testing and demonstrating video tools.
[SAFE]: The skill includes explicit 'Gates' and rules to ensure safe operation, such as mandating the use of the -t flag in ffmpeg to cap processing time and using deterministic seeding for all randomized values. This ensures that the agent's actions are predictable, seekable, and resource-bounded.

footage-effects