Demo Producer logo

Demo Producer

Community
yonatangross
demo-producer

Universal demo video creator for skills, agents, plugins, tutorials, CLI commands, and code walkthroughs. Generates scripts, storyboards, VHS terminal recordings, and Remotion video compositions with task-tracked production phases. Use when producing video showcases, marketing content, or terminal recordings.

Overview

Publisheryonatangross
Repositoryorchestkit
Skill namedemo-producer
Stars
298
Forks
35
Bundled files
17
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 17 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by yonatangross on GitHub. Read the source before you install it.

Installation

Install the Demo Producer AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/yonatangross/orchestkit/tree/main/plugins/ork/skills/demo-producer
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/yonatangross/orchestkit.git /tmp/orchestkit
mkdir -p .claude/skills
cp -r /tmp/orchestkit/plugins/ork/skills/demo-producer .claude/skills/demo-producer
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Demo Producer in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Demo Producer on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Demo Producer is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Demo Producer

Universal demo video creation for any content type.

Tool versions (2026-Q2): VHS >= 0.11, Remotion >= 4.0.448. VHS 0.11 added:

  • Wait keyword — supersedes Sleep N with a condition-based wait (Wait "prompt>" blocks until the given string appears). Use Wait over Sleep for any prompt that can appear at variable times.
  • ScrollUp N / ScrollDown N — explicit scroll steps inside a recorded session; useful for demonstrating TUI apps and long less-style output.

Keep existing Sleep N only for fixed visual pacing where no event is available.

Quick Start

bash
demo-producer                    # Interactive mode - asks what to create
demo-producer skill explore      # Create demo for a skill
demo-producer plugin ork     # Create demo for a plugin
demo-producer tutorial "Building a REST API"  # Custom tutorial
demo-producer --live 4           # Live preview via dev --live (4-hour public funnel)

--live mode (M127 #1565)

--live N is a thin wrapper around dev --live N: it boots the dev stack behind portless --funnel (publicly exposed via Tailscale Funnel), records an expiry of N hours into .claude/state/live-demos.jsonl, and prints a public URL plus <iframe> snippet for embedding in customer emails or docs pages.

bash
demo-producer --live 4

# Internally runs:
dev --live 4
# → public URL: https://app.your-tailnet.ts.net
#   iframe: <iframe src="..." />
#   expires: 2026-05-03T20:00Z

doctor warns about live demos older than 24 hours so they don't sprawl. Tear down with dev stop. Tailscale CLI must be installed (only required behind --live — not for static demo recording modes).

CRITICAL: Task Management is MANDATORY (CC 2.1.16)

BEFORE doing ANYTHING else, create tasks to track progress:

python
# 1. Create main task IMMEDIATELY
TaskCreate(
  subject="Demo Producer: {topic}",
  description="Creating polished demo video for content showcase",
  activeForm="Producing demo for {topic}"
)

# 2. Create subtasks for each production phase
TaskCreate(subject="Content analysis", activeForm="Analyzing content for demo")
TaskCreate(subject="Script & storyboard", activeForm="Generating script and storyboard")
TaskCreate(subject="Record & compose", activeForm="Recording terminal and composing video")
TaskCreate(subject="Final output", activeForm="Rendering final video outputs")

# 3. Set dependencies for sequential phases
TaskUpdate(taskId="3", addBlockedBy=["2"])
TaskUpdate(taskId="4", addBlockedBy=["3"])
TaskUpdate(taskId="5", addBlockedBy=["4"])

# 4. Update status as you progress
TaskUpdate(taskId="2", status="in_progress")  # When starting
TaskUpdate(taskId="2", status="completed")    # When done

Supported Content Types

TypeSourceExample
skillskills/{name}/SKILL.mddemo-producer skill commit
agentagents/{name}.mddemo-producer agent debug-investigator
pluginplugins/{name}/plugin.jsondemo-producer plugin ork
marketplaceMarketplace install flowdemo-producer marketplace ork
tutorialCustom descriptiondemo-producer tutorial "Git workflow"
cliAny CLI tooldemo-producer cli "npm create vite"
codeCode walkthroughdemo-producer code src/api/auth.ts

Interactive Flow

When invoked without arguments, asks 4 questions:

Question 1: Content Type

What type of demo do you want to create?

○ Skill - OrchestKit skill showcase
○ Agent - AI agent demonstration
○ Plugin - Plugin installation/features
○ Tutorial - Custom coding tutorial
○ CLI Tool - Command-line tool demo
○ Code Walkthrough - Explain existing code

Question 2: Format

What format(s) do you need?

☑ Horizontal (16:9) - YouTube, Twitter
☑ Vertical (9:16) - TikTok, Reels, Shorts
☐ Square (1:1) - Instagram, LinkedIn

Question 3: Style

What style fits your content?

○ Quick Demo (6-10s) - Fast showcase, single feature
○ Standard Demo (15-25s) - Full workflow, multiple steps
○ Tutorial (30-60s) - Detailed explanation, code examples
○ Cinematic (60s+) - Story-driven, high polish
○ Scrapbook (15-35s) - Warm paper, fast cuts, social proof collage (Anthropic style)

Question 4: Audio

Audio preferences?

○ Music Only - Subtle ambient background
○ Music + SFX - Background + success sounds
○ Silent - No audio

Pipeline Architecture

Load details: Read("references/demo-pipeline.md") for the full pipeline diagram, generation commands, and output structure.

Content Detector -> Content Analyzer -> Script Generator -> Terminal Script -> VHS Recorder -> Remotion Composer -> Final Outputs (horizontal/vertical/square).

Finish line and budget

Finish line. Done means: every requested format variant is rendered, written to the output directory, verified with ffprobe (non-zero duration, expected resolution per variant), and the completion PushNotification is sent.

Budget: at most 2 render retries per video, 3 format variants, and 45 minutes wall clock per variant (135 minutes total); stop and report at the finish line or the first cap, whichever comes first, naming the variants still unrendered.

Template System

Four template architectures for different demo styles. Load Read("references/template-system.md") for detailed configuration and the SkillDemoConfig interface.

TemplateUse CaseDurationKey Feature
TriTerminalRaceComplexity comparisons15-20s3-panel split, color-coded difficulty
ProgressiveZoomTutorials, walkthroughs20-30sZoom transitions, layered reveals
SplitThenMergeBefore/after, transformations15-25sSplit screen -> unified merge
ScrapbookDemoProduct launches, social proof15-35sWarm paper aesthetic, fast cuts

Content type templates (skill, agent, plugin, tutorial, cli, code) are mapped in Read("references/skill-category-mapping.md").

Remotion Composition

Load details: Read("references/remotion-composition.md") for folder structure, adding new compositions, and format variant prefixes.

Compositions organized under Production/ by format (Landscape, Vertical, Square) and skill category.

Customization Options

Visual Themes

  • Dark mode (default): Dark backgrounds, neon accents
  • Light mode: Clean whites, subtle shadows
  • Terminal: Pure terminal aesthetic
  • Cinematic: High contrast, dramatic lighting
  • Scrapbook: Warm paper (#F0F0E8), serif typography, fast cuts, mixed media collage

Audio Presets

  • Ambient: Subtle background, no SFX
  • Tech: Electronic beats, UI sounds
  • Corporate: Professional, clean
  • Energetic: Upbeat, fast-paced

PushNotification on Completion (CC 2.1.110+)

VHS recording and Remotion rendering commonly take 10–45 min per video, and full multi-variant productions (landscape + vertical + square) can exceed an hour. After all output files are written and verified, call PushNotification so the user knows the video is ready for review.

python
PushNotification(
  message=f"ork:demo-producer complete — {topic}: {variants_count} variants · {output_dir}",
  status="proactive"
)

Full rule: Read("${CLAUDE_PLUGIN_ROOT}/skills/chain-patterns/rules/push-notification-on-completion.md").

Best Practices

  1. Keep it focused - One feature/concept per video
  2. Show, don't tell - Demonstrate actual usage
  3. Use real data - Show actual command outputs
  4. Include context - Brief setup before the demo
  5. End with CTA - Always include install command

Terminal Simulation Patterns

Load details: Read("references/terminal-simulation.md") for TypeScript patterns: pinned header + scrolling content, agent color palette, and task spinner animation.

Slop Avoidance

Load details: Read("rules/slop-avoidance.md") for text density rules, timing compression, common slop patterns, and hook styles.

Core rule: If content doesn't earn its screen time, cut it.

Rules Quick Reference

RuleImpactWhat It Covers
analyzer-patterns (load rules/analyzer-patterns.md)MEDIUMFrontmatter parsing, phase detection, example extraction
production-pipeline (load rules/production-pipeline.md)HIGHPre-production, storyboarding, recording, VHS, manim
production-composition (load rules/production-composition.md)HIGHRemotion composition, audio mixing, thumbnails, captions
slop-avoidance (load rules/slop-avoidance.md)HIGHText density, timing compression, hook styles

Related Skills

  • video-production: Full video production pipeline (recording, composition, audio, pacing)

References

Load on demand with Read("references/<file>"):

FileContent
template-system.mdTemplate architecture and SkillDemoConfig interface
content-types.mdDetailed content type specs
format-selection.mdPlatform requirements and multi-format support
script-generation.mdScript templates and generation patterns
demo-pipeline.mdPipeline architecture, generation commands, output structure
remotion-composition.mdRemotion folder structure and composition guide
skill-category-mapping.mdSkill category mapping and content type templates

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Demo Producer AI skill do?

Universal demo video creator for skills, agents, plugins, tutorials, CLI commands, and code walkthroughs. Generates scripts, storyboards, VHS terminal recordings, and Remotion video compositions with task-tracked production phases. Use when producing video showcases, marketing content, or terminal recordings.

Why use Demo Producer on TypingMind?

Because you install it once and use it with any model. Demo Producer is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Demo Producer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/yonatangross/orchestkit/tree/main/plugins/ork/skills/demo-producer. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Demo Producer?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Demo Producer?

As many as you like. As long as a model supports skills, you can use Demo Producer with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Demo Producer AI skill free?

Yes. It is published on GitHub by yonatangross under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇