技能 happyhorse-1-0
🎬

happyhorse-1-0

中风险 ⚙️ 外部命令🌐 网络访问

Generate AI Videos with HappyHorse 1.0

也可从以下获取: doany-ai,agentspace-so

Create professional text-to-video content with HappyHorse 1.0, the top-ranked model on the Artificial Analysis Video Arena. Generate native 1080p videos with synchronized audio in a single pass.

支持: Claude Codex Code(CC)
⚠️ 63
1

下载技能 ZIP

2

在 Claude 中上传

前往 设置 → 功能 → 技能 → 上传技能

3

开启并开始使用

测试它

正在使用“happyhorse-1-0”。 A barista in a black apron pulls a single espresso shot, steam rising into the morning sun, rich crema slowly forming

预期结果:

A 5-second 1080p vertical video showing a close-up handheld shot of the espresso extraction process with warm cafe ambience and the hiss of the steam wand

正在使用“happyhorse-1-0”。 Shot 1: Medium close-up. A woman in a navy trench coat enters a rain-slick neon-lit Tokyo alley. Shot 2: Same woman in same navy trench, now under the awning of a ramen shop

预期结果:

A two-shot sequence with consistent character appearance: first shot shows her entering the alley under rain, second shows her under the awning with warm interior glow and gentle rain sounds

安全审计

中风险
v1 • 5/29/2026

This is a legitimate AI video generation skill for RunComfy's HappyHorse 1.0 model. The static analyzer flagged 86 potential issues, but evaluation reveals most are false positives from markdown code blocks being misidentified as executable code. The skill legitimately invokes the RunComfy CLI to call an external API for video generation. No malicious intent detected. Security documentation is comprehensive.

1
已扫描文件
196
分析行数
5
发现项
1
审计总数
中风险问题 (1)
External CLI Command Invocation
The skill invokes the RunComfy CLI (`runcomfy run`) to execute video generation requests. While this is legitimate functionality for an AI video service, external command execution is a risk factor that warrants warning for users.
低风险问题 (2)
Static Analyzer Pattern False Positives
The static analyzer flagged 66 'external_commands' patterns (Ruby/shell backtick execution) and 9 'weak cryptographic algorithm' flags, but these are all in markdown code blocks showing bash command examples. The SKILL.md file is documentation, not executable code. No actual shell execution occurs from this file.
Path Traversal Pattern False Positive
The static analyzer flagged path traversal sequences at lines 180-182, but these are legitimate CLI operations: the CLI downloads generated video outputs and polls an API endpoint. These are expected filesystem operations for a video generation tool.

风险因素

审计者: claude

质量评分

38
架构
100
可维护性
87
内容
30
社区
56
安全
91
规范符合性

你能构建什么

Create Multi-Shot Brand Stories

Generate consistent character appearances across multiple video segments for branded content campaigns. Ideal for advertisers who need a single character to appear consistently across a series of short-form videos.

Produce Talking-Head Explainers

Create educational or promotional videos with voiceover and ambient audio generated simultaneously with the video. The in-pass audio feature eliminates the need for separate audio production workflows.

Generate Multilingual Short-Form Ads

Produce platform-native vertical videos (9:16) with localized prompts in 6 supported languages. Perfect for social media campaigns targeting international audiences without script quality degradation.

试试这些提示

Basic Cinematic Prompt
Wide shot. [subject description with motion]. [camera movement]. [lighting direction and quality]. [additional visual details].
Multi-Shot Character Consistency
Shot 1: [scene setup with character description including clothing, appearance details]. Shot 2: Same [character description from Shot 1], [new action or location with consistent appearance description].
Vertical Platform Content
9:16 vertical short. [subject action with visual details]. [camera style for vertical]. [audio direction with ambient or dialogue cues].
Complex Multi-Element Scene
[Shot type]. [Primary subject with specific appearance and clothing]. [Secondary element or environment]. [Temporal progression of action]. [Cinematic lens and lighting specs]. [Audio direction]

最佳实践

  • Describe motion over time with temporal verbs rather than static descriptions. Say 'A woman turns from the window, walks two paces' instead of 'A woman drinking coffee'
  • Front-load camera and shot direction at the start of your prompt. Phrases like 'Wide shot.', 'Tracking shot.', or 'Locked tripod, low angle.' work as real directives
  • Restate character appearance anchors in each shot when generating multi-shot content. HappyHorse holds the look but needs the anchor description repeated

避免

  • Avoid static-frame descriptions with no temporal verbs - motion will be vague and unfocused
  • Do not pile up conflicting style directions in a single prompt - conflicting directions cancel each other out
  • Keep prompts under 2,500 characters - longer prompts cause quality degradation in generation

常见问题

What video resolution does HappyHorse 1.0 support?
HappyHorse 1.0 supports native 1080p output as well as 720p for faster, cheaper generations. Resolution is set via the 'resolution' parameter in the API call.
How long can the generated videos be?
Video duration can be set from 3 to 15 seconds per generation. For longer narratives, you should segment into multiple multi-shot prompts and stitch the results together.
Can I use external audio to drive lip-sync?
No, this text-to-video endpoint does not support external audio input. Audio is generated in-pass with the video. For audio-driven lip-sync, use Wan 2.7 or Seedance 2.0 Pro instead.
What languages are supported for prompts?
Prompts can be written in 6 languages: Chinese, English, Japanese, Korean, German, and French. There is no script quality drop between languages.
How do I maintain character consistency across multiple shots?
Include a full appearance anchor (clothing, colors, accessories) in each shot description. For example: 'Shot 1: A woman in a red wool coat, blue scarf. Shot 2: Same woman in red coat with blue scarf, now walking through a doorway.'
What happens if I use an unsupported aspect ratio?
Unsupported aspect ratios will result in a 422 error. HappyHorse 1.0 only supports five ratios: 16:9, 9:16, 1:1, 4:3, and 3:4.

开发者详情

文件结构

📄 SKILL.md