ai-video-generation

To generate AI video from text or image prompts, the skill routes your request to the correct RunComfy CLI endpoint based on whether you want text-to-video, image-to-video, or video extension, and executes a one-shot CLI invocation with the required JSON input body.

The best way to extend a Veo video clip is by using the skill's model-router to target the specific Veo video-extend endpoint, which maintains consistent motion and identity while calling the RunComfy CLI with your duration settings.

Model routing for AI video generation works by automatically selecting the best RunComfy endpoint based on your intent, identifying whether you need text-to-video, image-to-video, or video extension, and applying the correct vendor and model configuration.

Yes, you can create dialog-driven clips with lip sync by providing an audio_url in the JSON input body, which routes through the RunComfy CLI to generate video clips that match audio behavior and dialog.

No, you do not need to manually configure RunComfy CLI endpoints because the skill ships with a model-router by intent that automatically handles the correct vendor, model, and endpoint selection for text-to-video, image-to-video, and Veo extend tasks.