Seedance Video Generation

Prerequisites

pip install 'volcengine-python-sdk[ark]'

Features

• Text-to-video: Generate videos based on text prompts
• Image-to-video: Generate videos based on the first frame image
• Custom parameters: Support setting duration, aspect ratio, watermark, etc.
• Task management: Create tasks, query status, and auto download results

Usage

1. Quick video generation (text-to-video)

cd ~/.openclaw/workspace/skills/seedance-video
python3 scripts/generate_video.py "A cute cat playing on the grass" --wait -o cat.mp4

2. Custom parameters

python3 scripts/generate_video.py "Seaside at sunset" \
  --duration 10 \
  --ratio 16:9 \
  --wait \
  -o sunset.mp4

3. Image-to-video

python3 scripts/generate_video.py "Make the cat move" \
  --image-url https://example.com/cat.jpg \
  --duration 5 \
  --wait

4. Create task only (no waiting)

python3 scripts/generate_video.py "City under the starry sky" -o starry.mp4
# Return task ID

python3 scripts/generate_video.py --status <task_id> --wait -o starry.mp4

Parameter Description

prompt

-o, --output

-m, --model

-d, --duration

-r, --ratio

--watermark

--return-last-frame

--image-url

--wait

--status

API Key Configuration

ARK_API_KEY

SEEDANCE_API_KEY

Configuration Method (Recommended)

1. Copy the configuration template:

cp .canghe-skills/.env.example .canghe-skills/.env

1. Edit the

   .canghe-skills/.env

   file and fill in your API Key:

ARK_API_KEY=your-actual-api-key-here

Or use environment variables

export ARK_API_KEY="your-api-key"
# Or
export SEEDANCE_API_KEY="your-api-key"

Loading Priority

1. System environment variables (

   process.env

   )

2. .canghe-skills/.env

   in the current directory

3. ~/.canghe-skills/.env

   in the user home directory

Prompt Optimization Suggestions

1. Specific description: Describe details such as scene, subject, action, environment, etc.
2. Clear style: Specify the photography style (such as "cinematic", "documentary style")
3. Light description: Specify lighting conditions (such as "golden hour", "soft natural light")
4. Lens language: Describe camera movement (such as "slow push-in", "stabilizer shooting")

Example Prompts

A golden retriever running in an autumn park, golden fallen leaves falling, afternoon sunlight passing through the leaves, cinematic lens, stabilizer shooting, 4K quality

A robot walking on a neon street in a future city, cyberpunk style, rainy night, reflections, wide-angle lens, cinematic color tone

Python API Usage

from volcenginesdkarkruntime import Ark
from scripts.generate_video import create_video_task, wait_for_video

# Initialize client
client = Ark(
    base_url="https://ark.cn-beijing.volces.com/api/v3",
    api_key="your-api-key"
)

# Create video task
result = create_video_task(
    client=client,
    model="doubao-seedance-1-5-pro-251215",
    prompt="A bird flying in the sky",
    duration=5
)

task_id = result["task_id"]

# Wait and get result
final_status = wait_for_video(client, task_id)
video_url = final_status["video_url"]

Notes

• Asynchronous generation: Video generation is an asynchronous process, usually taking 30-60 seconds
• Task retention time: Task data is only retained for 24 hours
• Rate limiting: Pay attention to the RPM and concurrency limits of your account
• Video ratio: It is recommended that the video aspect ratio is close to the ratio of the first frame image

Task Status

• queued

  : In queue

• running

  : Generating

• succeeded

  : Succeeded

• failed

  : Failed