<< All versions

Skill v1.0.0

currentAutomated scan100/100
revfactory/harness-100/content-atomization
──Details
PublishedOctober 13, 2026 at 11:21 PM
Content Hashsha256:1b837d67ec4ed342...
Git SHA8e8d35c6a191
──Files
Files (1 file, 5.0 KB)
SKILL.md5.0 KBactive
SKILL.md · 115 lines · 5.0 KB

version: "1.0.0" name: content-atomization description: "Content atomization skill used by the source analyst and presentation builder. Provides methodology for breaking a single piece of content into its smallest units and recombining them for maximum conversion efficiency. Use for 'content decomposition,' 'atomization,' 'key extraction,' 'recombination,' and similar requests."


Content Atomization — Content Atomization Methodology

Specialist content decomposition knowledge used by the source-analyst and presentation-builder agents when analyzing sources and developing conversion strategies.

Why Atomization?

Repurposing is not "rewriting the same thing." It's breaking the source into atomic units and recombining them for each platform. A 2,000-word blog post can yield 30+ independent content pieces.

Content Atom Classification System

Atom TypeDefinitionExtraction ExampleConvertible Formats
ClaimA verifiable assertion"Remote work increases productivity by 13%"Tweet, card, slide
DataNumbers, statistics, ratios"2024 AI market size: $184B"Infographic, chart, tweet
StoryPerson/event-based narrative"Company A's transformation case"Thread, carousel, video script
ProcessStep-by-step methodology"5-step OKR setting method"Guide, carousel, checklist
QuoteCitations, aphorisms"'Failure is the mother of success'"Text card, tweet, slide
QuestionThought-provoking questions"What time is your team most productive?"Poll, comment prompt, opening
ContrastComparisons"Traditional vs. Agile methodology"Comparison table, carousel, infographic
DefinitionConcept definitions"What is OKR?"Educational content, slide, short-form

Source Analysis Process: The MINE Framework

M — Map (Structure Mapping)

Visually map the source's structure:

Source Content
├── Introduction
│ ├── [Claim] Core assertion
│ └── [Data] Supporting data
├── Body 1
│ ├── [Definition] Concept definition
│ ├── [Process] Methodology
│ └── [Story] Case study
├── Body 2
│ ├── [Contrast] Comparative analysis
│ ├── [Data] Statistics
│ └── [Quote] Expert citation
└── Conclusion
├── [Claim] Final assertion
└── [Question] Question for the reader

I — Identify (Core Atom Identification)

Assign a reuse value score to each atom:

CriterionWeightDescription
Independence30%Does it make sense without context?
Emotional Response25%Does it trigger surprise/empathy/outrage?
Shareability25%Is it information worth passing along?
Visualizability20%Can it be expressed as a graph/image?

N — Negotiate (Conversion Priority)

Set conversion priority by reuse value score:

  • Top 20%: Convert for all platforms (core messages)
  • Middle 50%: Convert for 2–3 suitable platforms
  • Bottom 30%: Keep in blog/long-form only, skip short-form conversion

E — Extend (Augmentation)

Add elements not in the source but needed for conversion:

  • Visual assets: If the source is text-only, add charts/diagrams
  • Personalization: If the source is objective, add experience/opinion layers
  • Action items: If the source is analytical, add actionable steps
  • Updates: If the source has outdated data, replace with current information

Presentation Conversion Formula

Slide Count Estimation

Recommended slides = (Presentation time in minutes × 1.5) + 3
- +3: Cover (1) + Table of Contents (1) + Q&A (1)
- ×1.5: ~40 seconds per slide

Slide Type Templates

TypeText AmountVisual RatioUse Case
TitleTitle + subtitle100% background imageSection dividers
Key Message1 sentence70%Core claim emphasis
DataLabels only80% chart/graphStatistics, comparisons
ProcessStep names only70% flowchartMethodology, steps
Comparison2-column table50%A vs B
QuoteQuote + sourceSpeaker imageAuthority/emotion
Story3–4 sentences50% related imageCases, anecdotes

The 1:10:30 Rule

  • 1 core message: The single claim the entire presentation delivers
  • 10 or fewer supporting points: Arguments backing the core message
  • 30 minutes or less: Attention limit (beyond this, interactive breaks are mandatory)

Conversion Efficiency Metrics

Source TypeExpected AtomsExpected Derivative Content
2,000-word blog15–258 tweets + 2 carousels + 15 slides + 3 short-form
5,000-word report30–503 blogs + 5 carousels + 25 slides + 2 newsletters
30-min interview20–305 clips + 10 tweets + 1 blog + 3 carousels
60-min talk40–603 blogs + 5 carousels + 15 tweets + 2 newsletters
All versions