AI Agent Hub
Back to skills
Vidu Q2 Image-to-Video icon

Vidu Q2 Image-to-Video

Design & Media Updated 2026.08.29

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @user_f95f21b8/dlazy-viduq2-i2v into your AI assistant.

About this skill

Problem

Static images cannot directly express camera motion, character action, or scene change. Turning an image into a dynamic video usually requires more than calling a model: it also involves dLazy authentication, local media upload, async task polling, and error handling. This skill wraps the Vidu Q2 image-to-video CLI workflow into a more stable flow, so developers or agents can run dlazy viduq2-i2v, submit prompts, image paths, and generation parameters, then receive video results or task states.

How It Works and Limits

  • Authentication: Requests require a dLazy API key. The key can be saved through the CLI into the user config directory, or supplied per invocation via the DLAZY_API_KEY environment variable. Keys are scoped to a dLazy organization and can be rotated or revoked from the dashboard.
  • Inference and storage: The skill is a thin client over dLazy's hosted API. Prompts and parameters are sent to the dLazy API for inference; local image, video, or audio paths are uploaded to dLazy media storage; generated output URLs are typically hosted on files.dlazy.com.
  • Parameter passing: In addition to standard flags, the CLI supports pipe-style references such as @N, @N.path, @stdin, and @stdin:path, which are useful for chaining upstream outputs in agent workflows.
  • Async tasks: When using --no-wait, the response may not include the final output. It may instead return a task.generateId and status, and the task state should be polled with dlazy status --wait.
  • Error boundaries: Common errors include 401 missing or invalid API key, 501 missing required parameters such as --prompt, 502 local file read failures, 503 insufficient balance or server errors, and 504 async generation failures, for example when a prompt violates safety policy.
  • Usage limits: It is not an offline inference tool, and local media files are uploaded to dLazy services. If the response returns insufficient_balance, credits must be topped up first. If it returns unauthorized, a valid API key must be configured before resuming the task.

Use Cases

  • Turn a product hero image into short video footage by adding camera motion and detail changes with a prompt.
  • Pass an upstream-generated image into `dlazy viduq2-i2v` in an agent workflow and forward the video URL downstream.
  • Submit image-to-video jobs with `--no-wait`, then poll task status using `dlazy status --wait`.
  • Diagnose `unauthorized` or `insufficient_balance` responses and guide users to configure a key or recharge.

Best For

  • Visual designers who need to turn design drafts, product images, or posters into short video previews.
  • Agent engineers who need to call image-to-video APIs in automation and handle task polling.
  • Operations planners producing short-form video assets who need a unified cloud generation service.
  • Platform engineers who need to debug dLazy CLI auth, insufficient balance, and local file read errors.