shortvideo

Create videos using ShortVideo API. Supports product-to-video, image-to-ad-video, and replicate-video. Use this skill when users want to: generate product videos, create ad videos from images, or replicate existing videos. Also trigger on: 生成视频, 产品视频, 视频生成, 制作视频, video generation, create video, product video, ad video.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "shortvideo" with this command: npx skills add thecur/shortvideo-skills

ShortVideo Creator

Create videos using ShortVideo backend API with multiple task types.

Authentication Setup (Required for First Use)

ShortVideo requires API credentials. Configure via environment variables:

Method 1: Claude Code Config

Add to ~/.claude/settings.json:

{
  "env": {
    "SHORTVIDEO_BASE_URL": "https://api.shortvideo.ai",
    "SHORTVIDEO_API_KEY": "your-api-key-here"
  }
}

Method 2: OpenClaw Config

Add to ~/.openclaw/openclaw.json:

{
  "skills": {
    "entries": {
      "shortvideo": {
        "env": {
          "SHORTVIDEO_BASE_URL": "https://api.shortvideo.ai",
          "SHORTVIDEO_API_KEY": "your-api-key-here"
        }
      }
    }
  }
}

Method 3: Shell Environment Variable

Add to ~/.zshrc (or ~/.bashrc):

export SHORTVIDEO_BASE_URL="https://api.shortvideo.ai"
export SHORTVIDEO_API_KEY="your-api-key-here"

Then reload: source ~/.zshrc

Note: When executing scripts, if environment variables are not found, try running with source ~/.zshrc && python3 scripts/...


Subcommands

product-to-video - Generate product video from image

Generate a marketing video from a single product image using Sora2 AI.

Trigger: product-to-video, 产品视频, product video

source ~/.zshrc && python3 scripts/product-to-video.py \
  --product-name "<product_name>" \
  --image <image_path> \
  --aspect-ratio <16:9|9:16> \
  --duration 12 \
  [--product-info "<description>"] \
  [--no-poll]

Parameters:

ParameterRequiredValid ValuesDescription
--product-nameYes-Product name
--imageYeslocal/OSS/URLProduct image path
--aspect-ratioYes16:9, 9:16Video ratio
--durationYes12Duration (only 12s supported)
--product-infoNo-Product description
--no-pollNo-Disable auto polling

Credit Cost: 100 (fixed)

Example:

source ~/.zshrc && python3 scripts/product-to-video.py \
  --product-name "Premium Headphones" \
  --image https://example.com/product.jpg \
  --aspect-ratio 16:9 \
  --duration 12

image-to-ad-video - Create ad video from images

Create advertisement videos from 1-7 product images.

Trigger: image-to-ad-video, 广告视频, ad video, image to video

source ~/.zshrc && python3 scripts/image-to-ad-video.py \
  --images <image1> <image2> ... \
  --duration <8|15|30|60> \
  --aspect-ratio <16:9|9:16|1:1> \
  [--prompt "<text>"] \
  [--no-poll]

Parameters:

ParameterRequiredValid ValuesDescription
--imagesYes1-7 pathsImage paths (local/OSS/URL)
--durationYes8, 15, 30, 60Duration in seconds
--aspect-ratioYes16:9, 9:16, 1:1Video ratio
--promptNomax 2000 charsStyle prompt
--no-pollNo-Disable auto polling

Credit Cost: 30 per second

Example:

source ~/.zshrc && python3 scripts/image-to-ad-video.py \
  --images d2mm4m9addr0008000a0.png \
  --duration 15 \
  --aspect-ratio 16:9

replicate-video - Replicate video with new images

Replicate an existing video style with new product/model images.

Trigger: replicate-video, 视频复刻, video replication

source ~/.zshrc && python3 scripts/replicate-video.py \
  --video <video_path> \
  --aspect-ratio <16:9|9:16> \
  --resolution <540p|720p|1080p> \
  [--product-images <img1> <img2> ...] \
  [--model-images <img1> <img2> ...] \
  [--prompt "<text>"] \
  [--remove-audio] \
  [--no-poll]

Parameters:

ParameterRequiredValid ValuesDescription
--videoYeslocal/OSS/URLSource video (5-300s)
--aspect-ratioYes16:9, 9:16Video ratio
--resolutionYes540p, 720p, 1080pOutput resolution
--product-imagesNo*1-7 pathsProduct images
--model-imagesNo*1-7 pathsModel images
--promptNomax 2000 charsStyle prompt
--remove-audioNo-Remove audio
--no-pollNo-Disable auto polling

*At least one image (product or model) is required.

Credit Cost: 9-15 per second (by resolution)

Example:

source ~/.zshrc && python3 scripts/replicate-video.py \
  --video template.mp4 \
  --aspect-ratio 16:9 \
  --resolution 1080p \
  --product-images product.jpg

poll-videos - Poll for video results

Poll for video generation results status.

Trigger: poll-videos, 查询视频, check video status

source ~/.zshrc && python3 scripts/poll-videos.py --video-ids <id1> <id2> [options]

Parameters:

ParameterRequiredDescription
--video-idsYesVideo IDs to check
--onceNoSingle check without polling
--max-attemptsNoMax polling attempts (default: 60)
--intervalNoPolling interval in seconds (default: 10)

Video Status Codes:

StatusDescription
0Pending
1Processing
2Completed
3Failed

Example:

# Continuous polling
source ~/.zshrc && python3 scripts/poll-videos.py --video-ids video_abc123

# Single check
source ~/.zshrc && python3 scripts/poll-videos.py --video-ids video_abc123 --once

File Path Types

All file parameters support:

TypeExampleBehavior
Local file/Users/xxx/file.jpgUpload to OSS
OSS pathd2mm4m9addr0008000a0.pngUse directly
URLhttps://example.com/file.jpgDownload → Upload

Supported File Types

TypeExtensionsMax Size
Images.jpg, .jpeg, .png, .gif, .webp, .bmp100MB
Videos.mp4, .mov, .avi, .mkv, .webm100MB

API Response Format

// Success
{"code": 0, "data": {...}}

// Failure
{"code": 1, "info": "error message"}

Reference Documentation

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

Aavegotchi Renderer Bypass

Render Aavegotchi assets by deriving renderer hashes from Goldsky Base core data and calling POST /api/renderer/batch on www.aavegotchi.com. Use when the use...

Registry SourceRecently Updated
General

Toutiao User Profile API

Call GET /api/toutiao/get-user-detail/v1 for Toutiao User Profile through JustOneAPI with userId.

Registry SourceRecently Updated
General

Toutiao Search API

Call 2 search versions for Toutiao App Keyword Search through JustOneAPI with keyword.

Registry SourceRecently Updated
General

Compaction UI Enhancements

Background memory compaction with auto-trigger, chat summary paragraph, configurable threshold, model selector, settings tab, and result storage for OpenClaw...

Registry SourceRecently Updated