git clone https://github.com/digitalsamba/claude-code-video-toolkit.git
cd claude-code-video-toolkit

# Install Python dependencies (optional, for AI tools)
python3 -m pip install -r tools/requirements.txt

# Open in Claude Code
claude

# First-time setup (cloud GPU, voice, storage)
/setup

# Create your first video
/video

# Or render example immediately (no API keys needed)
cd examples/hello-world
npm install
npm run render

cd projects/my-project
npm run studio

/generate-voiceover
# Or directly:
python tools/voiceover.py --script script.md --output audio.mp3

/record-demo
# Generates MP4 from automated browser session

claude-code-video-toolkit/
├── .claude/               # Claude Code skills and commands
│   ├── skills/           # Skill definitions (remotion, elevenlabs, etc.)
│   └── commands/         # Command definitions (/video, /setup, etc.)
├── projects/             # Video projects (gitignored)
├── templates/            # Project templates
│   ├── sprint-review/
│   └── product-demo/
├── brands/               # Brand profiles
│   ├── default/
│   └── digital-samba/
├── lib/                  # Shared libraries
│   ├── transitions/      # Scene transition effects
│   ├── project/          # Project management system
│   └── components/       # Reusable Remotion components
├── tools/                # Python AI tools
│   ├── voiceover.py      # Generate AI speech
│   ├── qwen_edit.py      # AI image editing
│   ├── acestep.py        # AI music generation
│   └── ltx2.py           # AI video generation
└── examples/             # Example projects

/video
# Select "Create new project"
# Choose template (e.g., sprint-review-v2)
# Specify brand profile
# Answer planning questions

{
  "intent": "Show iOS app login screen with animation",
  "visualType": "demo-recording",
  "duration": 8,
  "assets": {
    "video": {
      "status": "needed",
      "path": null
    }
  }
}

# Record demo with Playwright
/record-demo

# Generate AI voiceover
python tools/voiceover.py \
  --script projects/my-project/audio/script.md \
  --output projects/my-project/audio/voiceover.mp3 \
  --voice $ELEVENLABS_VOICE_ID

# Generate AI image
python tools/qwen_edit.py \
  --prompt "futuristic tech background, dark blue" \
  --output projects/my-project/assets/background.png

# Generate AI music
python tools/acestep.py \
  --prompt "upbeat electronic, tech demo" \
  --duration 60 \
  --output projects/my-project/audio/music.mp3

cd projects/my-project
npm run studio
# Opens http://localhost:3000 for live preview

cd projects/my-project
npm run render
# Outputs to out/video.mp4

// src/scenes/IntroScene.tsx
import { AbsoluteFill, useCurrentFrame } from 'remotion';
import { interpolate } from 'remotion';

export const IntroScene: React.FC<{ title: string }> = ({ title }) => {
  const frame = useCurrentFrame();
  const opacity = interpolate(frame, [0, 30], [0, 1]);
  
  return (
    <AbsoluteFill style={{ 
      backgroundColor: '#0a0e27',
      justifyContent: 'center',
      alignItems: 'center'
    }}>
      <h1 style={{ 
        opacity,
        fontSize: 72,
        color: '#00d4ff'
      }}>
        {title}
      </h1>
    </AbsoluteFill>
  );
};

import { Series } from 'remotion';
import { glitch, fade } from '../../lib/transitions';
import { IntroScene } from './scenes/IntroScene';
import { DemoScene } from './scenes/DemoScene';

export const Video: React.FC = () => (
  <Series>
    <Series.Sequence durationInFrames={90}>
      <IntroScene title="My Product" />
    </Series.Sequence>
    <Series.Transition
      presentation={glitch()}
      timing={{ durationInFrames: 15 }}
    />
    <Series.Sequence durationInFrames={240}>
      <DemoScene videoSrc="assets/demo.mp4" />
    </Series.Sequence>
  </Series>
);

import { 
  glitch,        // Digital distortion
  rgbSplit,      // Chromatic aberration
  zoomBlur,      // Radial motion blur
  lightLeak,     // Lens flare
  clockWipe,     // Radial sweep
  pixelate,      // Mosaic dissolution
  checkerboard,  // Grid reveal (9 patterns)
  slide,         // Official Remotion
  fade,          // Official Remotion
  wipe,          // Official Remotion
  flip           // Official Remotion
} from '../../lib/transitions';

# tools/voiceover.py
import os
from elevenlabs.client import ElevenLabs

client = ElevenLabs(api_key=os.getenv("ELEVENLABS_API_KEY"))

audio = client.generate(
    text="Welcome to our product demo",
    voice="EXAVITQu4vr4xnSDxMaL",  # Voice ID
    model="eleven_multilingual_v2"
)

with open("voiceover.mp3", "wb") as f:
    for chunk in audio:
        f.write(chunk)

# Deploy to Modal
modal deploy tools/qwen_tts_modal.py

# Generate speech
python tools/voiceover.py \
  --script script.md \
  --output audio.mp3 \
  --provider qwen

# script.md

In this sprint, we shipped three major features.

[pause:1.0]

First, we rebuilt the authentication system with biometric support.

[emphasis]This was a big win for security.[/emphasis]

[pause:0.5]

Let's see it in action.

# tools/qwen_edit.py (uses FLUX.2 internally)
python tools/qwen_edit.py \
  --prompt "cyberpunk city at night, neon lights, rain" \
  --output background.png \
  --width 1920 \
  --height 1080

python tools/qwen_edit.py \
  --input photo.jpg \
  --prompt "add sunglasses and a leather jacket" \
  --output edited.png

# tools/acestep.py
python tools/acestep.py \
  --prompt "upbeat electronic music for tech demo" \
  --duration 60 \
  --output music.mp3 \
  --lyrics "optional verse lyrics here"

/brand
# Select "Create new brand"

brands/my-brand/
├── brand.json      # Visual identity
├── voice.json      # Voice settings
└── assets/
    ├── logo.png
    └── background.jpg

{
  "name": "My Brand",
  "colors": {
    "primary": "#00d4ff",
    "secondary": "#ff006e",
    "background": "#0a0e27",
    "text": "#ffffff"
  },
  "fonts": {
    "heading": "Inter",
    "body": "Inter"
  },
  "typography": {
    "headingSize": 72,
    "bodySize": 24
  }
}

{
  "provider": "elevenlabs",
  "voiceId": "EXAVITQu4vr4xnSDxMaL",
  "stability": 0.5,
  "similarityBoost": 0.75,
  "model": "eleven_multilingual_v2"
}

# Install Modal
pip install modal

# Set up token
modal token new

# Deploy AI tools
modal deploy tools/qwen_tts_modal.py
modal deploy tools/qwen_edit_modal.py
modal deploy tools/acestep_modal.py

# Configure RunPod endpoint
export RUNPOD_API_KEY=your_key
export RUNPOD_ENDPOINT_ID=your_endpoint

# Use with tools
python tools/voiceover.py --provider qwen --runpod

cat projects/my-project/project.json | grep phase

import { Video, AbsoluteFill } from 'remotion';

export const OverlayScene: React.FC = () => (
  <AbsoluteFill>
    <Video src="demo.mp4" />
    <AbsoluteFill style={{ 
      justifyContent: 'flex-end',
      padding: 40 
    }}>
      <h2 style={{ color: 'white', fontSize: 48 }}>
        Feature Demo
      </h2>
    </AbsoluteFill>
  </AbsoluteFill>
);

import { useCurrentFrame } from 'remotion';
import { interpolate } from 'remotion';

export const StatsScene: React.FC = () => {
  const frame = useCurrentFrame();
  const value = Math.floor(interpolate(frame, [0, 60], [0, 1247]));
  
  return (
    <AbsoluteFill style={{ 
      justifyContent: 'center',
      alignItems: 'center'
    }}>
      <div style={{ fontSize: 120, fontWeight: 'bold' }}>
        {value.toLocaleString()}
      </div>
      <div style={{ fontSize: 36, marginTop: 20 }}>
        GitHub Stars
      </div>
    </AbsoluteFill>
  );
};

import { Audio, useCurrentFrame } from 'remotion';

export const NarratedScene: React.FC = () => {
  const frame = useCurrentFrame();
  const fps = 30;
  
  // Show text at specific timestamps
  const showText = frame > fps * 2.5 && frame < fps * 5;
  
  return (
    <AbsoluteFill>
      <Audio src="voiceover.mp3" />
      {showText && <h2>This appears during narration</h2>}
    </AbsoluteFill>
  );
};

# ElevenLabs (optional)
export ELEVENLABS_API_KEY=your_key
export ELEVENLABS_VOICE_ID=voice_id

# Modal (recommended for AI tools)
export MODAL_TOKEN_ID=your_token
export MODAL_TOKEN_SECRET=your_secret

# RunPod (alternative)
export RUNPOD_API_KEY=your_key
export RUNPOD_ENDPOINT_ID=endpoint_id

# Cloudflare R2 (optional, for file transfer)
export R2_ACCOUNT_ID=account_id
export R2_ACCESS_KEY_ID=access_key
export R2_SECRET_ACCESS_KEY=secret_key
export R2_BUCKET_NAME=bucket_name

# Increase Node memory
export NODE_OPTIONS="--max-old-space-size=8192"
npm run render

# Reinstall Modal
pip install --upgrade modal

# Check authentication
modal token new

# Verify deployment
modal app list

# Ensure @remotion/transitions is installed
cd projects/my-project
npm install @remotion/transitions