generate-video

Generate Video Skill

A collection of skills responsible for automatically generating product explanation videos.

---

Overview

This skill is used internally by the /generate-video command. It executes the flow of codebase analysis → scenario proposal → parallel generation.

Feature Details

Feature	Details
Best Practices	See references/best-practices.md
Codebase Analysis	See references/analyzer.md
Scenario Planning	See references/planner.md
Parallel Scene Generation	See references/generator.md
Visual Effects Library	See references/visual-effects.md
AI Image Generation	See references/image-generator.md
Image Quality Assessment	See references/image-quality-check.md

Prerequisites

• Remotion is set up (/remotion-setup)
• Node.js 18+
• (Optional) GOOGLE_AI_API_KEY - for AI image generation

/generate-video Flow

/generate-video
    │
    ├─[Step 1] Analysis (analyzer.md)
    │   ├─ Framework detection
    │   ├─ Key feature detection
    │   ├─ UI component detection
    │   └─ Project asset analysis (Plans.md, CHANGELOG, etc.)
    │
    ├─[Step 2] Scenario Proposal (planner.md)
    │   ├─ Auto-determine video type
    │   ├─ Scene composition proposal
    │   └─ User confirmation
    │
    ├─[Step 2.5] Asset Generation (image-generator.md) ← NEW
    │   ├─ Determine if assets are needed (intro, CTA, etc.)
    │   ├─ Generate 2 images with Nano Banana Pro
    │   ├─ Claude assesses quality (image-quality-check.md)
    │   └─ OK → adopt / NG → regenerate (up to 3 times)
    │
    └─[Step 3] Parallel Generation (generator.md)
        ├─ Parallel scene generation (Task tool)
        ├─ Integration + transitions
        └─ Final rendering

Execution Steps

1. User runs /generate-video
2. Verify Remotion setup
3. Analyze codebase with analyzer.md
4. Propose scenario with planner.md + user confirmation
5. Parallel generation with generator.md
6. Completion report

Video Types (by funnel stage)

Type	Funnel Stage	Approx. Length	Auto-detection Criteria	Core Structure
LP/Ad Teaser	Awareness → Interest	30-90s	New project	Pain → Result → CTA
Intro Demo	Interest → Consideration	2-3 min	UI changes detected	Complete 1 use case
Release Notes	Consideration → Conviction	1-3 min	CHANGELOG updated	Before/After focused
Architecture Overview	Conviction → Decision	5-30 min	Major structural changes	Production use + evidence
Onboarding	Retention & Adoption	30s - several min	First-time setup	Shortest path to Aha moment

Details: references/best-practices.md

Scene Templates

90-second Teaser (for LP/ads)

Time	Scene	Content
0-5s	Hook	Pain or desired outcome
5-15s	Problem+Promise	Target user and promise
15-55s	Workflow	Signature workflow
55-70s	Differentiator	Basis for differentiation
70-90s	CTA	Next step

3-minute Intro Demo (for consideration)

Time	Scene	Content
0-10s	Hook	Conclusion + pain
10-30s	UseCase	Use case declaration
30-140s	Demo	Full walkthrough on actual screens
140-170s	Objection	Address one common concern
170-180s	CTA	Call to action

Common Scenes

Scene	Recommended Duration	Content
Intro	3-5s	Logo + tagline
Feature Demo	10-30s	Playwright capture
Architecture Diagram	10-20s	Mermaid → animation
CTA	3-5s	URL + contact

Detailed templates: ${CLAUDE_SKILL_DIR}/references/best-practices.md

Audio Sync Rules (Important)

Strictly follow these rules for narrated videos:

Rule	Value
Audio start	Scene start + 30f (1-second wait)
Scene length	30f + audio length + 20f padding
Transition	15f (overlap with adjacent scenes)
Scene start calculation	Previous scene start + previous scene length - 15f

Pre-check: Verify audio length with ffprobe before designing scenes

Details: ${CLAUDE_SKILL_DIR}/references/generator.md

BGM Support

Item	Recommended Value
With narration	bgmVolume: 0.20 - 0.30
Without narration	bgmVolume: 0.50 - 0.80
File location	public/BGM/

Details: ${CLAUDE_SKILL_DIR}/references/generator.md

Subtitle Support

Rule	Value
Subtitle start	Same as audio start
Subtitle duration	Audio length + 10f
Font	Base64 embedding recommended

Details: ${CLAUDE_SKILL_DIR}/references/generator.md

Visual Effects Library

A collection of effects for impactful videos:

Effect	Use Case
GlitchText	Hook, titles
Particles	Background, CTA convergence
ScanLine	Analysis in-progress effect
ProgressBar	Parallel processing display
3D Parallax	Card display

Details: references/visual-effects.md

Notes

• If Remotion is not set up, guide users to /remotion-setup
• Number of parallel generations auto-adjusts based on scene count (max 5)
• Generated videos are output to the out/ directory
• AI-generated images are saved to out/assets/generated/
• When GOOGLE_AI_API_KEY is not set, image generation is skipped (existing assets or placeholders are used)