create-video

/create-video - Multi-Scene Video Creation Workflow

Create a complete educational video with multiple scenes, consistent styling, and proper concatenation.

Workflow Steps

Step 1: Video Planning (@viz-architect)

Gather high-level requirements:

1. Topic: Main subject of the video
2. Target Audience: Beginner / Intermediate / Advanced
3. Duration: Short (3-5 min) / Medium (5-10 min) / Long (10-15 min)
4. Style Preferences: Colors, pacing, formality

Step 2: Scene Breakdown (@scene-designer)

Create the video structure:

# Video: [Title]

**Total Duration**: ~X minutes
**Scene Count**: Y scenes

## Scene List

| # | Scene Name | Duration | Concept | Camera |
|---|------------|----------|---------|--------|
| 1 | TitleScene | 20s | Video title | static |
| 2 | MotivationScene | 60s | Why this matters | static |
| 3 | ... | ... | ... | ... |

## Detailed Scene Plans

### Scene 1: TitleScene (20s)
[Timeline and technical requirements]

### Scene 2: MotivationScene (60s)
[Timeline and technical requirements]

...

Step 3: Shared Configuration

Create a config module used by all scenes:

# video_config.py
from manim import *

# Video Title
VIDEO_TITLE = "Understanding [Topic]"

# Semantic Color Palette (consistent across all scenes)
COLORS = {
    "PRIMARY": BLUE,        # Main concept (e.g., the function)
    "SECONDARY": ORANGE,    # Supporting concept (e.g., derivatives)
    "POSITIVE": GREEN,      # Success, correct answers
    "EMPHASIS": PURPLE,     # Highlights, key insights
    "MUTED": GRAY,          # De-emphasized, background
}

# Typography
FONTS = {
    "TITLE": 52,
    "SECTION": 42,
    "BODY": 28,
    "SMALL": 24,
    "FORMULA": 38,
}

# Timing Constants
TIMING = {
    "FAST": 0.5,
    "NORMAL": 1.0,
    "SLOW": 2.0,
    "DRAMATIC": 3.0,
    "BRIEF": 0.5,
    "STANDARD": 1.0,
    "EXTENDED": 2.0,
    "COMPLEX": 4.0,
    "CRITICAL": 6.0,
}

# Base Scene Class
class VideoScene(Scene):
    """Base class for all scenes in this video"""

    def setup(self):
        self.tracked = []

    def track(self, *mobjects):
        for m in mobjects:
            self.tracked.append(m)
            self.add(m)
        return mobjects[0] if len(mobjects) == 1 else mobjects

    def cleanup(self, keep=None):
        keep = keep or []
        to_remove = [m for m in self.tracked if m not in keep]
        if to_remove:
            self.play(*[FadeOut(m) for m in to_remove])
        self.tracked = list(keep)

    def section_title(self, text):
        """Standard section title animation"""
        title = Text(text, font_size=FONTS["SECTION"])
        self.play(Write(title), run_time=TIMING["NORMAL"])
        self.wait(TIMING["BRIEF"])
        self.play(title.animate.to_edge(UP, buff=0.5))
        return title