> ## Documentation Index
> Fetch the complete documentation index at: https://docs.shanone.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Wan (Alibaba)

> Generate video from text or images using Alibaba's Wan model via DashScope

# Wan (Alibaba)

Shanone's Wan integration gives your agent 4 tools for generating video from a text prompt or a source image, and for tracking those (asynchronous) generation jobs through Alibaba Cloud's DashScope API. It's a small, focused integration rather than a broad platform API.

## Getting Started

<Steps>
  <Step title="Generate a DashScope API key">
    In the [Alibaba Cloud DashScope console](https://dashscope.console.aliyun.com/), create an API key with access to the Wan video generation models.
  </Step>

  <Step title="Connect the API key in Shanone">
    Ask your agent to do something with Wan (e.g. "generate a video of a paper airplane flying through a city"), or add the API key proactively from **Integrations** in the Shanone dashboard.
  </Step>

  <Step title="Retry your request">
    Once connected, `shanone_execute_tool` calls for `wan_*` tools will succeed.
  </Step>
</Steps>

## Available Tools

Shanone provides **4 tools** for Wan, organized into these categories:

<AccordionGroup>
  <Accordion title="Video generation">
    `wan_text_to_video`, `wan_image_to_video`
  </Accordion>

  <Accordion title="Task tracking">
    `wan_get_task`, `wan_wait_for_task`
  </Accordion>
</AccordionGroup>

## Common Use Cases

<CardGroup cols={2}>
  <Card title="Text-to-video generation" icon="video">
    Turn a scene description into a short video clip for a storyboard or social post
  </Card>

  <Card title="Image animation" icon="image">
    Animate a static product photo or illustration into a short motion clip
  </Card>

  <Card title="Batch content generation" icon="layers">
    Queue several generation tasks and poll them for completion in a review pipeline
  </Card>

  <Card title="Creative exploration" icon="sparkles">
    Generate multiple variations from the same prompt to pick the best result
  </Card>
</CardGroup>

## Troubleshooting

<AccordionGroup>
  <Accordion title="wan_text_to_video or wan_image_to_video doesn't return a video immediately">
    Video generation is asynchronous — both tools return a task ID that you should pass to `wan_wait_for_task` (or poll manually with `wan_get_task`) until the job completes.
  </Accordion>

  <Accordion title="wan_image_to_video fails on the source image">
    Check `shanone_get_tool_schema` for `wan_image_to_video` — the source image typically needs to be a publicly accessible URL rather than raw binary data, and must meet DashScope's format/size requirements.
  </Accordion>
</AccordionGroup>