ai-content-pipeline

安装 belt CLI 技能： npx skills add belt-sh/cli

AI 内容流水线

通过 inference.sh CLI 构建多步骤内容创作流水线。

快速开始

需要 inference.sh CLI（belt）。安装说明

belt login

# Simple pipeline: Generate image -> Animate to video
belt app run falai/flux-dev --input '{"prompt": "portrait of a woman smiling"}' > image.json
belt app run falai/wan-2-5-i2v --input '{"image": "<url-from-previous>", "prompt": "slow camera push-in"}'

流水线模式

模式 1：图像 -> 视频 -> 音频

[FLUX Image] -> [Wan 2.5 Video] -> [Foley Sound]

模式 2：脚本 -> 语音 -> 虚拟形象

[LLM Script] -> [Kokoro TTS] -> [OmniHuman Avatar]

模式 3：研究 -> 内容 -> 分发

[Tavily Search] -> [Claude Summary] -> [FLUX Visual] -> [Twitter Post]

完整工作流

YouTube 短视频流水线

根据主题创作完整的短视频。

# 1. Generate script with Claude
belt app run openrouter/claude-sonnet-45 --input '{
  "prompt": "Write a 30-second script about the future of AI. Make it engaging and conversational. Just the script, no stage directions."
}' > script.json

# 2. Generate voiceover with Kokoro
belt app run falai/kokoro-tts --input '{
  "prompt": "<script-text>",
  "voice": "af_sarah"
}' > voice.json

# 3. Generate background image with FLUX
belt app run falai/flux-dev --input '{
  "prompt": "Futuristic city skyline at sunset, cyberpunk aesthetic, 4K wallpaper"
}' > background.json

# 4. Animate image to video with Wan
belt app run falai/wan-2-5-i2v --input '{
  "image": "<background-url>",
  "prompt": "slow camera pan across cityscape, subtle movement"
}' > video.json

# 5. Add captions (manually or with another tool)

# 6. Merge video with audio
belt app run infsh/video-audio-merger --input '{
  "video_file": "<video-url>",
  "audio_file": "<voice-url>"
}'

讲解视频流水线

创建展示内容的 AI 虚拟形象。

# 1. Write the script
belt app run openrouter/claude-sonnet-45 --input '{
  "prompt": "Write a 1-minute explainer script about quantum computing for beginners."
}' > script.json

# 2. Generate speech
belt app run falai/kokoro-tts --input '{
  "prompt": "<script>",
  "voice": "am_michael"
}' > speech.json

# 3. Generate or use a portrait image
belt app run falai/flux-dev --input '{
  "prompt": "Professional headshot of a friendly tech presenter, neutral background, looking at camera"
}' > portrait.json

# 4. Create talking head video
belt app run bytedance/omnihuman-1-5 --input '{
  "image": "<portrait-url>",
  "audio": "<speech-url>"
}' > talking_head.json

产品演示流水线

创建产品展示视频。

# 1. Generate product image
belt app run falai/flux-dev --input '{
  "prompt": "Sleek wireless earbuds on white surface, studio lighting, product photography"
}' > product.json

# 2. Animate product reveal
belt app run falai/wan-2-5-i2v --input '{
  "image": "<product-url>",
  "prompt": "slow 360 rotation, smooth motion"
}' > product_video.json

# 3. Upscale video quality
belt app run falai/topaz-video-upscaler --input '{
  "video": "<product-video-url>"
}' > upscaled.json

# 4. Add background music
belt app run infsh/video-audio-merger --input '{
  "video_file": "<upscaled-url>",
  "audio_file": "https://your-music.mp3"
}'

博客转视频流水线

将文字内容转换为视频格式。

# 1. Summarize blog post
belt app run openrouter/claude-haiku-45 --input '{
  "prompt": "Summarize this blog post into 5 key points for a video script: <blog-content>"
}' > summary.json

# 2. Generate images for each point
for i in 1 2 3 4 5; do
  belt app run falai/flux-dev --input "{
    \"prompt\": \"Visual representing point $i: <point-text>\"
  }" > "image_$i.json"
done

# 3. Animate each image
for i in 1 2 3 4 5; do
  belt app run falai/wan-2-5-i2v --input "{
    \"image\": \"<image-$i-url>\",
    \"prompt\": \"subtle camera motion\"
  }" > "video_$i.json"
done

# 4. Generate voiceover
belt app run falai/kokoro-tts --input '{
  "prompt": "<full-script>",
  "voice": "bf_emma",
  "language": "british-english"
}' > narration.json

# 5. Merge all clips
belt app run infsh/media-merger --input '{
  "media_files": [
    {"file": "<video1>", "transition_type": "crossfade"},
    {"file": "<video2>", "transition_type": "crossfade"},
    {"file": "<video3>", "transition_type": "crossfade"},
    {"file": "<video4>", "transition_type": "crossfade"},
    {"file": "<video5>", "transition_type": "crossfade"}
  ]
}' > merged.json

belt app run infsh/video-audio-merger --input '{
  "video_file": "<merged-url>",
  "audio_file": "<narration-url>"
}'

流水线构建模块

内容生成

步骤	应用	用途
脚本	openrouter/claude-sonnet-45	撰写内容
研究	tavily/search-assistant	收集信息
摘要	openrouter/claude-haiku-45	精简内容

视觉素材

步骤	应用	用途
图像	falai/flux-dev	生成图像
图像	google/imagen-3	图像生成替代方案
升级分辨率	falai/topaz-image-upscaler	提升质量

动画

步骤	应用	用途
I2V	falai/wan-2-5-i2v	让图像动起来
T2V	google/veo-3-1-fast	根据文本生成
虚拟形象	bytedance/omnihuman-1-5	讲解视频

音频

步骤	应用	用途
TTS	falai/kokoro-tts	语音旁白
音乐	infsh/ai-music	背景音乐
拟音	infsh/mmaudio	音效

后期制作

步骤	应用	用途
升级分辨率	falai/topaz-video-upscaler	提升视频质量
合并	infsh/media-merger	合并媒体
字幕	infsh/caption-videos	添加字幕

最佳实践

1. 先规划流水线 - 运行前梳理每个步骤
2. 保存中间结果 - 存储输出以便迭代
3. 选择合适的质量 - 草稿使用快速模型，最终成品使用高质量模型
4. 匹配分辨率 - 全程保持宽高比一致
5. 测试每个步骤 - 验证输出后再继续

相关技能

# Video generation models
npx skills add inference-sh/skills@ai-video-generation

# Image generation
npx skills add inference-sh/skills@ai-image-generation

# Text-to-speech
npx skills add inference-sh/skills@text-to-speech

# LLM models for scripts
npx skills add inference-sh/skills@llm-models

# Full platform skill
npx skills add inference-sh/skills@infsh-cli

浏览所有应用：belt app list

文档

• 内容流水线示例 - 官方流水线指南
• 构建工作流 - 工作流最佳实践