PromptHub 图片与视频生成

What it does

Lets a Skill-capable agent call the PromptHub API to generate images and videos. The agent can discover available models, submit asynchronous generation jobs, poll their status, and return the final image or video URL. Local reference images, videos, and audio are supported.

Usage profile

Suitable tasks

Ask an agent to generate images or videos from text, use reference images for image or image-to-video generation, submit long-running media jobs, and check model capabilities or task results.

Input and output

Input: a prompt, model ID, aspect ratio/duration/quality parameters, and optional reference assets. Output: a task ID, processing status, error details, and the completed image or video URL. Image and video requests use asynchronous jobs instead of holding the request open.

Permission signals

Set PROMPTHUB_API_KEY in the environment that runs the agent and allow the Node.js scripts to access the PromptHub API. Generation consumes PromptHub points; local reference assets are uploaded to PromptHub for the task.

Boundaries

This is an API client Skill, not a local model. It requires Node.js 18+ and an agent that supports SKILL.md; model availability, parameter limits, generation time, and pricing come from the live API.