Aliyun Vidu Video

What it does

When your agent uses it

• Generating videos with DashScope Vidu models
• Implementing text-to-video
• Image-to-video (first frame)
• Keyframe-to-video (first+last frame)

Example prompts

• “/aliyun-vidu-video”

Requirements

• Python 3
• A credential in DASHSCOPE_API_KEY
• A credential in API_KEY

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

1. 1Confirm user intent: text-to-video, image-to-video, keyframe, or reference-to-video.
2. 2Select the appropriate model name based on capability and quality tier (Q3 pro/turbo or Q2).
3. 3Prepare input: prompt and/or media array with correct types and valid public URLs.
4. 4Set resolution, size, and duration parameters.
5. 5Create async task and poll for results (15s interval recommended).
6. 6Download and save generated video before URL expiration (24 hours).

What it can do on your machine

Read from SKILL.md and the folder at commit 1818263. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 1 file in scripts/ (Python), which the agent can run.

  Shell commands in SKILL.md call:

  • python

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • dashscope.aliyuncs.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • DASHSCOPE_API_KEY
  • API_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Aliyun Vidu Video loads about 2.7k tokens when it runs, and up to ~4.7k if it reads all its reference files. Until then it costs about 61 tokens; SKILL.md has 785 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~61

When it runs · the whole SKILL.md, loaded when a task matches

~2.7k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~4.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from cinience/alicloud-skills at commit 1818263, republished under its MIT licence (© cinience). 785 words, ~2,696 tokens.

name

aliyun-vidu-video

description

Use when generating videos with DashScope Vidu models. Use when implementing text-to-video, image-to-video (first frame), keyframe-to-video (first+last frame), or reference-to-video generation via the video-synthesis async API.

mkdir -p output/aliyun-vidu-video
python -m py_compile skills/ai/video/aliyun-vidu-video/scripts/generate_vidu_video.py && echo "py_compile_ok" > output/aliyun-vidu-video/validate.txt

POST https://dashscope.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis

import os
import json
import time
import requests

API_KEY = os.getenv("DASHSCOPE_API_KEY")
BASE_URL = "https://dashscope.aliyuncs.com/api/v1"

def create_vidu_task(req: dict) -> str:
    """Create a Vidu video generation task and return task_id."""
    payload = {
        "model": req["model"],
        "input": {},
        "parameters": {
            "resolution": req.get("resolution", "720P"),
            "duration": req.get("duration", 5),
        },
    }
    if req.get("prompt"):
        payload["input"]["prompt"] = req["prompt"]
    if req.get("media"):
        payload["input"]["media"] = req["media"]
    if req.get("size"):
        payload["parameters"]["size"] = req["size"]
    if req.get("audio") is not None:
        payload["parameters"]["audio"] = req["audio"]
    if req.get("watermark") is not None:
        payload["parameters"]["watermark"] = req["watermark"]
    if req.get("seed") is not None:
        payload["parameters"]["seed"] = req["seed"]

    resp = requests.post(
        f"{BASE_URL}/services/aigc/video-generation/video-synthesis",
        headers={
            "Authorization": f"Bearer {API_KEY}",
            "Content-Type": "application/json",
            "X-DashScope-Async": "enable",
        },
        json=payload,
    )
    resp.raise_for_status()
    data = resp.json()
    return data["output"]["task_id"]


def poll_task(task_id: str, interval: int = 15) -> dict:
    """Poll until task completes. Returns final response."""
    while True:
        resp = requests.get(
            f"{BASE_URL}/tasks/{task_id}",
            headers={"Authorization": f"Bearer {API_KEY}"},
        )
        resp.raise_for_status()
        data = resp.json()
        status = data["output"]["task_status"]
        if status in ("SUCCEEDED", "FAILED", "CANCELED"):
            return data
        time.sleep(interval)

# Text-to-video
task_id = create_vidu_task({
    "model": "vidu/viduq3-turbo_text2video",
    "prompt": "A cat running under moonlight",
    "resolution": "540P",
    "size": "960*528",
    "duration": 5,
})

# Image-to-video (first frame)
task_id = create_vidu_task({
    "model": "vidu/viduq3-pro_img2video",
    "prompt": "Camera slowly pans upward",
    "media": [{"type": "image", "url": "https://example.com/image.jpg"}],
    "resolution": "720P",
    "duration": 5,
})

# Keyframe-to-video (first + last frame)
task_id = create_vidu_task({
    "model": "vidu/viduq3-turbo_start-end2video",
    "prompt": "A cat jumps from windowsill to sofa",
    "media": [
        {"type": "image", "url": "https://example.com/first.png"},
        {"type": "image", "url": "https://example.com/last.png"},
    ],
    "resolution": "540P",
    "duration": 5,
})

# Reference-to-video
task_id = create_vidu_task({
    "model": "vidu/viduq2_reference2video",
    "prompt": "Man playing guitar in a cafe",
    "media": [
        {"type": "image", "url": "https://example.com/ref1.jpg"},
        {"type": "image", "url": "https://example.com/ref2.jpg"},
    ],
    "resolution": "720P",
    "size": "1280*720",
    "duration": 5,
})

© cinience, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (scripts, references) in skills/ai/video/aliyun-vidu-video of cinience/alicloud-skills.

• SKILL.md
• references/api_reference.md
• references/sources.md
• scripts/generate_vidu_video.py

Open the folder on GitHubat commit 1818263