← Back to Skills Marketplace
krisyejh

HappyHorse Video Generation and Editting

by krisyejh · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
51
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install happyhorse-video-gen-edit
Description
Video Generation and Editing with HappyHorse models. Supports text2video, image2video (first-frame based), reference2video (multi-image character fusion), an...
README (SKILL.md)

HappyHorse Video Models

HappyHorse Video Models, created by Alibaba Group, are video generation and editing models. This skill integrates with HappyHorse Model APIs on ModelStudio (Bailian-Alibaba Model Service Platform).

Model Overview

Model Capabilities Resolution Duration
happyhorse-1.0-t2v Text to Video 720P, 1080P 3-15s
happyhorse-1.0-i2v First-frame Image to Video 720P, 1080P 3-15s
happyhorse-1.0-r2v Reference images to video, Multi-character fusion 720P, 1080P 3-15s
happyhorse-1.0-video-edit Video editing, Style transfer, Reference-based edit 720P, 1080P 3-15s (output)

Text to Video Generation (happyhorse-1.0-t2v)

Generate videos from text prompts.

text2video task-submit

python3 {baseDir}/scripts/happyhorse-magic.py text2video-gen --prompt "一座由硬纸板和瓶盖搭建的微型城市，在夜晚焕发出生机" --resolution "720P" --ratio "16:9" --duration 5
python3 {baseDir}/scripts/happyhorse-magic.py text2video-gen --prompt "一只可爱的小猫将军站在悬崖上，远处是壮丽的雪山" --resolution "1080P" --ratio "16:9" --duration 10
python3 {baseDir}/scripts/happyhorse-magic.py text2video-gen --prompt "海浪拍打岩石的慢镜头" --resolution "720P" --ratio "9:16" --duration 8 --no-watermark

Options

  • --prompt: Text prompt for video generation (required, max 5000 non-Chinese chars or 2500 Chinese chars)
  • --resolution: Video resolution - 720P or 1080P (default: 1080P)
  • --ratio: Aspect ratio - 16:9, 9:16, 1:1, 4:3, 3:4 (default: 16:9)
  • --duration: Video duration in seconds, 3-15 (default: 5)
  • --no-watermark: Disable "Happy Horse" watermark (default: enabled)
  • --seed: Random seed for reproducibility [0, 2147483647]

text2video tasks-get (round-robin)

python3 {baseDir}/scripts/happyhorse-magic.py text2video-get --task-id "\x3CTASK_ID_FROM_VIDEO_GEN>"

Image to Video Generation (happyhorse-1.0-i2v)

Generate video from a first-frame image. Output aspect ratio automatically follows the input image.

First-frame to Video

python3 {baseDir}/scripts/happyhorse-magic.py image2video-gen --prompt "一只猫在草地上奔跑" --first-frame "https://example.com/cat.png" --resolution "720P" --duration 5
python3 {baseDir}/scripts/happyhorse-magic.py image2video-gen --first-frame "https://example.com/landscape.jpg" --resolution "1080P" --duration 10
python3 {baseDir}/scripts/happyhorse-magic.py image2video-gen --prompt "花朵缓缓绽放" --first-frame "https://example.com/flower.png" --duration 8 --no-watermark

Options

  • --prompt: Text prompt for video generation (optional but recommended)
  • --first-frame: First frame image URL (required). Format: JPEG/JPG/PNG/WEBP, min 300px per side, max 10MB.
  • --resolution: Video resolution - 720P or 1080P (default: 1080P). Output ratio follows input image.
  • --duration: Video duration in seconds, 3-15 (default: 5)
  • --no-watermark: Disable "Happy Horse" watermark (default: enabled)
  • --seed: Random seed for reproducibility [0, 2147483647]

Note: There is no --ratio parameter. The output video aspect ratio automatically follows the input first-frame image.

image2video tasks-get (round-robin)

python3 {baseDir}/scripts/happyhorse-magic.py image2video-get --task-id "\x3CTASK_ID_FROM_VIDEO_GEN>"

Reference to Video Generation (happyhorse-1.0-r2v)

Generate video from reference images as character/object sources. Use "character1", "character2", etc. in the prompt to reference images by their array order.

Multi-character Reference

python3 {baseDir}/scripts/happyhorse-magic.py reference2video-gen --prompt "身着红色旗袍的女性character1，镜头以侧面中景勾勒旗袍修身剪裁" --reference-images "https://example.com/girl.jpg" --resolution "720P" --ratio "16:9" --duration 5
python3 {baseDir}/scripts/happyhorse-magic.py reference2video-gen --prompt "身着红色旗袍的女性character1，轻抬玉手展开折扇character2时流苏耳坠character3随头部转动轻盈摆动" --reference-images "https://example.com/girl.jpg" "https://example.com/fan.jpg" "https://example.com/earrings.jpg" --resolution "720P" --ratio "16:9" --duration 5
python3 {baseDir}/scripts/happyhorse-magic.py reference2video-gen --prompt "character1在海边散步，阳光洒在沙滩上" --reference-images "https://example.com/person.png" --resolution "1080P" --ratio "16:9" --duration 10

Options

  • --prompt: Text prompt with character references (required, max 5000 non-Chinese chars). Use "character1/character2/..." to reference images in array order.
  • --reference-images: Reference image URLs (required, 1-9 images). The 1st image = character1, 2nd = character2, etc.
  • --resolution: Video resolution - 720P or 1080P (default: 1080P)
  • --ratio: Aspect ratio - 16:9, 9:16, 1:1, 4:3, 3:4 (default: 16:9)
  • --duration: Video duration in seconds, 3-15 (default: 5)
  • --no-watermark: Disable "Happy Horse" watermark (default: enabled)
  • --seed: Random seed for reproducibility [0, 2147483647]

Note: Reference images should have short side >= 400px. Avoid low-resolution, blurry, or heavily compressed images.

reference2video tasks-get (round-robin)

python3 {baseDir}/scripts/happyhorse-magic.py reference2video-get --task-id "\x3CTASK_ID_FROM_VIDEO_GEN>"

Video Editing (happyhorse-1.0-video-edit)

Edit existing videos with text instructions and optional reference images. Supports style transfer, local replacement, and other editing tasks.

Instruction-based Editing

python3 {baseDir}/scripts/happyhorse-magic.py video-edit --prompt "将场景变成夜晚，添加霓虹灯效果" --video "https://example.com/video.mp4" --resolution "720P"
python3 {baseDir}/scripts/happyhorse-magic.py video-edit --prompt "为人物换上酷闪的衣服" --video "https://example.com/original.mp4"

Reference Image-based Editing

python3 {baseDir}/scripts/happyhorse-magic.py video-edit --prompt "让视频中的角色穿上图片中的条纹毛衣" --video "https://example.com/video.mp4" --reference-images "https://example.com/clothes.webp"
python3 {baseDir}/scripts/happyhorse-magic.py video-edit --prompt "将背景替换为参考图中的场景" --video "https://example.com/video.mp4" --reference-images "https://example.com/bg.jpg" --resolution "720P"

Options

  • --prompt: Editing instruction (required, max 5000 non-Chinese chars)
  • --video: Input video URL to edit (required, mp4/mov, 3-60s input, output max 15s)
  • --reference-images: Reference image URLs for editing (optional, 0-5 images)
  • --resolution: Output resolution - 720P or 1080P (default: 1080P)
  • --audio-setting: Audio handling - auto (default) or origin (keep original audio)
  • --no-watermark: Disable "Happy Horse" watermark (default: enabled)
  • --seed: Random seed for reproducibility [0, 2147483647]

Note: Output aspect ratio and duration follow the input video. No --ratio or --duration parameters. If input video > 15s, system auto-truncates to first 15s.

video-edit tasks-get (round-robin)

python3 {baseDir}/scripts/happyhorse-magic.py video-edit-get --task-id "\x3CTASK_ID_FROM_VIDEO_EDIT>"

Key Differences from Wan 2.7

Feature HappyHorse Wan 2.7
Watermark default Enabled ("Happy Horse") Disabled ("AI Generated")
prompt_extend Not supported Supported
negative_prompt Not supported Supported
seed parameter Supported (all models) Supported (all models)
I2V modes First-frame only First-frame, First+Last-frame, Video continuation
R2V reference type Images only (character1/2) Videos + Images (视频1/图片1)
R2V reference count 1-9 images Max 3 videos + 5 images
Video Edit input 3-60s, output max 15s 2-10s
Video Edit prompt Required Optional

Important Notes

  • Task ID Validity: Task IDs are valid for 24 hours
  • Video URL Validity: Generated video URLs are valid for 24 hours, download immediately
  • Content Review: Both input and output are subject to content safety review
  • Billing: Based on resolution (1080P > 720P) × duration (seconds)
  • Watermark: Default enabled with "Happy Horse" text; use --no-watermark to disable
Usage Guidance
This skill appears to do what it claims: it calls DashScope (HappyHorse) APIs and requires a DASHSCOPE_API_KEY. Before installing: 1) Only provide a DashScope API key you trust and that has limited scope; avoid using high-privilege or long-lived keys. 2) Be careful with local-file arguments: the included script will base64-embed and send local files to the external service, so do not point it at sensitive files (SSH keys, credentials, private documents). 3) Ensure the runtime environment has python3 and a safe requests installation (the script assumes the requests package). 4) If you enable autonomous agent invocation, consider limiting autonomy or auditing requests and outputs — autonomous agents could be instructed to upload arbitrary local files. 5) If you need higher assurance, validate the API endpoints and behavior by running the script in an isolated environment (container/VM) and inspect network traffic or server-side logs.
Capability Analysis
Type: OpenClaw Skill Name: happyhorse-video-gen-edit Version: 1.0.0 The skill provides a legitimate interface for Alibaba's DashScope HappyHorse video generation and editing models. The primary script, scripts/happyhorse-magic.py, implements standard API interactions with https://dashscope.aliyuncs.com, including support for local file uploads via base64 encoding and task status polling. No evidence of malicious intent, data exfiltration to unauthorized domains, or prompt injection was found.
Capability Tags
requires-sensitive-credentials
Capability Assessment
✓ Purpose & Capability
Name/description, referenced endpoints (dashscope.aliyuncs.com), and required environment variable (DASHSCOPE_API_KEY) align with the HappyHorse / DashScope AIGC video service; python3 requirement is expected for the provided script.
ℹ Instruction Scope
Runtime instructions tell the agent to run the included Python script which performs HTTP requests to dashscope.aliyuncs.com. The script will accept URLs or local file paths and will base64-embed local files (data: URIs) into the request body — this is necessary to upload local media but means arbitrary local files can be read and transmitted if the user or agent supplies those paths. The SKILL.md does not attempt to read unrelated system config or credentials.
✓ Install Mechanism
There is no install spec (instruction-only with an included script). Nothing is downloaded from unknown URLs and no installers run; the only runtime dependency is python3 (the script also uses the requests library, which is not installed by the skill).
✓ Credentials
Only one environment variable is required: DASHSCOPE_API_KEY (declared as primaryEnv) which is appropriate for authenticating to the DashScope API. No unrelated credentials or config paths are requested.
✓ Persistence & Privilege
The skill is not forced-always-enabled and does not request elevated or cross-skill config changes. Note: autonomous invocation is allowed by default (platform default) — combined with the ability to upload local files, this increases the practical blast radius if the agent is granted broad autonomy.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install happyhorse-video-gen-edit
  3. After installation, invoke the skill by name or use /happyhorse-video-gen-edit
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of happyhorse-video-gen-edit: video generation and editing with HappyHorse models. - Supports text-to-video, first-frame image-to-video, multi-image reference-to-video (character fusion), and video editing. - Provides command-line interfaces for all generation and editing tasks. - Allows configurable options for resolution (720P/1080P), aspect ratio, duration, and watermark control. - Reference-to-video supports up to 9 images for character fusion in prompts. - Video editing enables style transfer, background replacement, and reference-based modifications with flexible audio handling. - Includes clear differentiation from related models (Wan 2.7) in watermarking, prompt support, and reference image handling.
Metadata
Slug happyhorse-video-gen-edit
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is HappyHorse Video Generation and Editting?

Video Generation and Editing with HappyHorse models. Supports text2video, image2video (first-frame based), reference2video (multi-image character fusion), an... It is an AI Agent Skill for Claude Code / OpenClaw, with 51 downloads so far.

How do I install HappyHorse Video Generation and Editting?

Run "/install happyhorse-video-gen-edit" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is HappyHorse Video Generation and Editting free?

Yes, HappyHorse Video Generation and Editting is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does HappyHorse Video Generation and Editting support?

HappyHorse Video Generation and Editting is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created HappyHorse Video Generation and Editting?

It is built and maintained by krisyejh (@krisyejh); the current version is v1.0.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463942 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259883 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200739 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191401 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190375 · by oswalpalash

💬 Comments