imehr avatar

video-explainer

Use when creating explainer videos, video content for YouTube/TikTok/Instagram/LinkedIn, or when use

by imehr|Open Source

Video Explainer Skill

Create professional explainer videos across multiple platforms with consistent branding and intelligent automation.

Commands

Setup & Configuration

/video setup - Configure video_explainer path and initial settings

  • Prompts for video_explainer installation path
  • Validates the tool is accessible
  • Creates initial config.yaml

/video brands - List available brands /video brands create - Interactive brand creation wizard /video platforms - List available platform profiles /video styles - List available style overlays /video config [project] - Show/edit configuration

Project Management

/video new [name] - Create new video project

  • Without name: Interactive wizard (asks brand, platforms, style, languages)
  • With --quick: Skip wizard, use defaults
  • Creates project directory with config.json

/video list - List all video projects /video status [project] - Show project progress, outputs, versions

Pipeline Control

/video script [project] - Generate script from source document

  • Runs in foreground (user review expected)
  • Applies brand voice, platform pacing
  • Uses storytelling techniques from learned skills

/video scenes [project] - Generate scenes from script

  • Orchestrator decides parallel execution per aspect ratio
  • Applies brand visual system

/video render [project] - Render all outputs

  • --only main,tiktok - Specific outputs only
  • --lang en - Specific language only
  • --version v3 - Specific version number
  • Orchestrator runs renders in parallel background

Iteration & Feedback

/video feedback [project] "message" - Apply feedback and re-render

  • Creates new version automatically
  • Captures feedback for learning

/video compare [project] [output] [v1] [v2] - Side-by-side version review /video promote [project] [output] [version] - Promote version to final

Learning & Skills

/video learn "instruction" - Teach a new preference or technique

  • --from-url URL - Learn from web content
  • --from-file path - Learn from local file

/video memory - Show learning summary

  • --patterns - Discovered patterns
  • --preferences - Inferred preferences
  • --effectiveness - Technique scores

/video forget "technique" - Remove a learned technique /video prefer "preference" --always - Set strong preference

/video skills - List loaded technique skills /video skills sync - Pull updates from skill sources

Export

/video export [project] --zip - Package final outputs for delivery

Orchestrator Behavior

The intelligent orchestrator analyzes each task and decides:

  1. Foreground vs Background - User-interactive steps (script review) run foreground; heavy processing runs background
  2. Parallel vs Sequential - Independent tasks (different aspect ratios, languages) run in parallel
  3. Dependencies - Tracks what each step needs and schedules accordingly

Example execution plan:

/video render my-project (YouTube + TikTok, en/es)

1. [FOREGROUND] Script generation (user may review)
2. [BACKGROUND PARALLEL]
   - Scene generation (16:9)
   - Scene generation (9:16)
   - Voiceover (en)
   - Voiceover (es)
3. [BACKGROUND PARALLEL] (when dependencies ready)
   - Render YouTube-en
   - Render YouTube-es
   - Render TikTok-en
   - Render TikTok-es

Quality Gates

Every output passes through validation:

  • Concept coverage: Key concepts from source appear in script
  • Audio sync: Voiceover duration matches scenes, no awkward pauses
  • Narrative flow: Hook present, transitions smooth, CTA at end
  • Fact check: Claims supported by source material

Configuration Matrix

Three independent axes combine:

Output = Brand × Platform × Style
  • Brand: Voice, colors, typography, assets (comprehensive design system)
  • Platform: Aspect ratio, duration limits, pacing, safe zones
  • Style: Mood overlay (energetic, minimal, corporate)

File Locations

~/.config/video-explainer/
├── config.yaml           # Global settings
├── brands/               # Brand definitions
├── platforms/            # Platform profiles
├── styles/               # Style overlays
├── skills/               # Technique definitions
└── memory/               # Learning data

Sub-Skills

This skill automatically loads technique modules:

  • techniques/hooks.md - Opening techniques for engagement
  • techniques/narrative.md - Story structure patterns
  • techniques/platform-optimization.md - Platform-specific tips

Integration with video_explainer CLI

This skill wraps the video_explainer Python CLI tool:

# Under the hood, commands map to CLI calls:
/video script → video_explainer script
/video scenes → video_explainer scenes
/video render → video_explainer render

The orchestrator handles config translation, parallel execution, and output management.

Download Skill