Back to Skills

content-os

majiayu000
Updated 2 days ago
15 views
58
9
58
View on GitHub
Metadesign

About

Content OS is a master orchestrator skill that generates all content types from a single seed idea in forward mode or splits long-form content into short-form pieces in backward mode. It coordinates a pipeline of specialized skills for research, writing, quality review, and visual generation, applying full quality gates to long-form output and a quick accuracy pass to short-form content. Use it as a unified "produce everything" button for multi-format content creation workflows.

Quick Install

Claude Code

Recommended
Plugin CommandRecommended
/plugin add https://github.com/majiayu000/claude-skill-registry
Git CloneAlternative
git clone https://github.com/majiayu000/claude-skill-registry.git ~/.claude/skills/content-os

Copy and paste this command in Claude Code to install this skill

Documentation

Content OS: Multi-Format Content Orchestrator

The "produce everything" button. Give one seed idea → get all content types. Or give long-form content → get it split into short-form pieces.

Quick Start

Forward Mode (Seed → All Content)

User: "Content OS: Statins myth-busting for Indians"

Output:
├── Long-form (quality-passed)
│   ├── YouTube script (Hinglish)
│   ├── Newsletter (B2C - patients)
│   ├── Newsletter (B2B - doctors)
│   ├── Editorial
│   └── Blog post
├── Short-form (accuracy-checked)
│   ├── 5-10 tweets
│   ├── 1 thread
│   └── Carousel content
└── Visual
    ├── Instagram carousel slides
    └── Infographic concepts

Backward Mode (Long-form → Split)

User: "Content OS: [paste your blog/script/newsletter]"

Output:
├── 5-10 tweets (key points)
├── 1 thread (condensed narrative)
├── Carousel slides (visual summary)
└── Snippets (quotable sections)

How It Works

Mode Detection

  • Forward Mode: Input is a topic/idea (short text, question, or concept)
  • Backward Mode: Input is existing long-form content (>500 words)

Forward Mode Pipeline

PHASE 1: RESEARCH
│
├── PubMed MCP
│   └── Search for relevant papers, trials, guidelines
│
├── knowledge-pipeline (RAG)
│   └── Query AstraDB for ACC/ESC/ADA guidelines, textbooks
│
├── social-media-trends-research (optional)
│   └── Check trending angles, audience questions
│
└── OUTPUT: research-brief.md
    └── Synthesized knowledge with citations

PHASE 2: LONG-FORM CONTENT (Full Quality Pipeline)
│
├── youtube-script-master
│   └── Hinglish script → Quality Review → Final
│
├── cardiology-newsletter-writer
│   └── B2C newsletter → Quality Review → Final
│
├── medical-newsletter-writer
│   └── B2B newsletter → Quality Review → Final
│
├── cardiology-editorial
│   └── Editorial → Quality Review → Final
│
└── cardiology-writer
    └── Blog post → Quality Review → Final

PHASE 3: SHORT-FORM CONTENT (Quick Accuracy Pass)
│
├── x-post-creator-skill
│   └── 5-10 tweets → Accuracy Check → Final
│
├── twitter-longform-medical
│   └── Thread → Accuracy Check → Final
│
└── Extract carousel content from long-form

PHASE 4: VISUAL CONTENT
│
├── carousel-generator
│   └── Generate Instagram slides from key points
│
└── cardiology-visual-system
    └── Infographic concepts (if data-heavy)

PHASE 5: OUTPUT
│
└── Organized folder structure with all content

Backward Mode Pipeline

PHASE 1: ANALYZE
│
└── Parse long-form content
    ├── Extract key points
    ├── Identify data/statistics
    ├── Find quotable sections
    └── Determine topic/theme

PHASE 2: SPLIT (Quick Accuracy Pass)
│
├── Generate tweets (5-10)
│   └── One key point per tweet
│
├── Generate thread
│   └── Condensed narrative
│
├── Extract carousel content
│   └── Key points for slides
│
└── Create snippets
    └── Quotable sections

PHASE 3: VISUAL
│
└── carousel-generator
    └── Generate slides from extracted content

PHASE 4: OUTPUT
│
└── All short-form pieces organized

Quality Gates

Long-Form Quality Pipeline (FULL)

Each long-form piece goes through:

  1. scientific-critical-thinking

    • Evidence rigor check
    • Citation verification
    • Claim accuracy
    • Statistical interpretation
  2. peer-review

    • Methodology review
    • Logical consistency
    • Completeness check
    • Counter-argument consideration
  3. content-reflection

    • Pre-publish QA
    • Audience appropriateness
    • Clarity check
    • Structure review
  4. authentic-voice

    • Anti-AI pattern removal
    • Voice consistency
    • Natural language check

Short-Form Accuracy Pass (QUICK)

Each short-form piece gets:

  1. Data Interpretation Check
    • Are trial results stated correctly?
    • Are statistics accurately represented?
    • Is the study conclusion not misrepresented?
    • Are effect sizes/NNT/HR correctly stated?

This is a sanity check, not full review. User can iterate manually.

Skills Invoked

Research Skills

SkillPurpose
knowledge-pipelineRAG + PubMed synthesis
PubMed MCPDirect paper search
social-media-trends-researchTrending angles

Writing Skills

SkillContent TypeQuality Gate
youtube-script-masterYouTube script (Hinglish)Full
cardiology-newsletter-writerPatient newsletterFull
medical-newsletter-writerDoctor newsletterFull
cardiology-editorialEditorialFull
cardiology-writerBlog postFull
x-post-creator-skillTweetsQuick
twitter-longform-medicalThreadQuick

Quality Skills

SkillPurposeUsed For
scientific-critical-thinkingEvidence rigorLong-form
peer-reviewMethodology checkLong-form
content-reflectionPre-publish QALong-form
authentic-voiceAnti-AI cleanupLong-form

Visual Skills

SkillPurpose
carousel-generatorInstagram slides
cardiology-visual-systemInfographics

Repurposing Skills

SkillPurpose
cardiology-content-repurposerBackward mode splitting

Output Structure

/output/content-os/[topic-slug]/
├── research/
│   └── research-brief.md           # Foundation for all content
│
├── long-form/                       # Full quality pipeline
│   ├── youtube-script.md           ✓ Quality passed
│   ├── newsletter-b2c.md           ✓ Quality passed
│   ├── newsletter-b2b.md           ✓ Quality passed
│   ├── editorial.md                ✓ Quality passed
│   └── blog.md                     ✓ Quality passed
│
├── short-form/                      # Quick accuracy pass
│   ├── tweets.md                   ✓ Accuracy checked
│   ├── thread.md                   ✓ Accuracy checked
│   └── snippets.md                 ✓ Accuracy checked
│
├── visual/
│   ├── carousel/
│   │   └── slide-01.png...
│   └── infographic-concepts.md
│
└── summary.md                       # What was produced

Invocation Examples

Forward Mode

"Content OS: GLP-1 agonists cardiovascular benefits"
"Content OS: Statin myths for Indian patients"
"Content OS: When to get a CAC score"
"Content OS: SGLT2 inhibitors in heart failure"

Backward Mode

"Content OS: [paste your 2000-word blog post]"
"Content OS: [paste your YouTube script]"
"Content OS: [paste your newsletter]"

Configuration

What Gets Produced (Forward Mode)

Content TypeDefaultCan Skip
YouTube ScriptYesYes
Newsletter B2CYesYes
Newsletter B2BYesYes
EditorialYesYes
BlogYesYes
TweetsYesYes
ThreadYesYes
CarouselYesYes

Customization

"Content OS: Statins - only YouTube and tweets"
"Content OS: Heart failure - skip editorial"
"Content OS: CAC scoring - long-form only"

Integration with Existing System

Content OS orchestrates skills that already exist in your system. It doesn't replace them - it coordinates them.

You can still use individual skills directly:

  • youtube-script-master for just a script
  • x-post-creator-skill for just tweets
  • carousel-generator for just slides

Content OS is for when you want everything at once.

Notes

  • Long-form content takes longer due to quality pipeline
  • Short-form is faster (quick accuracy pass only)
  • Research phase runs once, shared by all content
  • Visual content generated from text output
  • All content uses same research foundation for consistency

Voice & Quality Standards

All content follows:

  • YouTube: Peter Attia depth + Hinglish (70% Hindi / 30% English)
  • Twitter/Writing: Eric Topol Ground Truths style
  • B2B (Doctors): JACC editorial voice
  • Anti-AI: No "It's important to note", no excessive hedging
  • Citations: Q1 journals, specific statistics, NNT/HR/CI when relevant

GitHub Repository

majiayu000/claude-skill-registry
Path: skills/content-os

Related Skills

content-collections

Meta

This skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.

View skill

creating-opencode-plugins

Meta

This skill provides the structure and API specifications for creating OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It offers implementation patterns for JavaScript/TypeScript modules that intercept and extend the AI assistant's lifecycle. Use it when you need to build event-driven plugins for monitoring, custom handling, or extending OpenCode's capabilities.

View skill

langchain

Meta

LangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.

View skill

cloudflare-turnstile

Meta

This skill provides comprehensive guidance for implementing Cloudflare Turnstile as a CAPTCHA-alternative bot protection system. It covers integration for forms, login pages, API endpoints, and frameworks like React/Next.js/Hono, while handling invisible challenges that maintain user experience. Use it when migrating from reCAPTCHA, debugging error codes, or implementing token validation and E2E tests.

View skill