videoagent-video-studio

You can generate AI videos from text prompts or images by providing a prompt or image reference, then specifying mode, duration, aspect ratio, and model. The skill routes requests through a hosted proxy to backends like MiniMax, Kling 3.0, or Veo, returning a videoUrl or jobId.

You can use seven text-to-video models without managing API keys: MiniMax, Kling 3.0, Google Veo, Grok, Hunyuan, Seedance, and PixVerse. Requests are routed through a hosted proxy, eliminating the need for multiple accounts or direct API key management.

Yes, you can use an image reference to create a short video. The skill supports image-to-video and reference-based generation workflows by accepting an imageUrl as input, animating your provided images through the selected backend model.

No, you do not need to configure a local proxy to generate AI videos. The skill uses a zero-setup workflow via a hosted proxy, though optional local or dev proxy configurations are available if needed for specific routing requirements.

Required inputs for text-to-video generation include the mode, prompt, duration, aspectRatio, and model. For image-to-video workflows, an imageUrl is also required. Token-based authentication is used to process these requests through the proxy.

Yes, there are optional rate-limiting mechanisms applied when routing AI video generation requests through the hosted proxy. This helps manage request frequency across the seven supported backends, ensuring stable videoUrl or jobId returns.