Framework Content Mixer

Structure:

1. Hook (Contrarian): "AI Tools That Never Fail? Here's What Nobody Tells You"
2. Setup: Common belief everyone holds
3. Challenge: Why that belief is wrong (with data)
4. Investigation: Your research methodology
5. Findings: What the data actually shows
6. Conclusion: New mental model

When to Use:

• Challenging industry best practices
• Presenting research findings
• Debunking myths with data
• Educational content for advanced audience

Example (Claude Code Context):

Hook: "Most developers think more documentation helps Claude Code. I analyzed 50 projects and found the opposite."

Structure:
• The conventional wisdom: More docs = better context
• Why it seems logical (information helps AI, right?)
• My experiment: 50 projects, tracked context quality vs. doc length
• Surprising finding: Projects with 3-5 focused CLAUDE.md files outperformed 20+ scattered READMEs
• The insight: Claude needs structure, not volume
• What to do instead: Hierarchical context architecture
• Real implementation from Personal-OS project

CTA: "Comment if you want my CLAUDE.md template"

Platforms: LinkedIn (detailed), YouTube (visual research), Twitter thread (key findings)

Framework #2: Contrarian + Fortune Teller

Formula: Challenge assumption + Predict future based on current trends

Structure:

1. Hook (Contrarian): "Everyone's rushing to X. Here's why that's wrong"
2. Current trend: What everyone is doing
3. Challenge: Why it will fail
4. Trend analysis: Data showing shift
5. Prediction: What's actually coming
6. Action items: How to prepare

When to Use:

• Industry trend commentary
• Strategic positioning content
• "Here's what's next" thought leadership
• Warning about hype cycles

Example:

Hook: "AI Skills That Always Fail? 7 approaches with shockingly high disappointment rates"

Structure:
• The hype: Everyone learning prompt engineering
• The problem: 70% give up within 3 months (data from surveys)
• Why they fail: Wrong expectations, no framework
• What's actually working: Framework-driven approaches
• Prediction: Frameworks will replace freestyle prompting
• How to position now: Learn proven patterns vs. experimenting
• Evidence: Companies hiring "framework specialists" not "prompt engineers"

CTA: "Get my framework library (link in bio)"

Platforms: LinkedIn (professional analysis), YouTube (trend deep-dive), Blog (comprehensive)

Framework #3: Experimenter + Fortune Teller

Formula: Personal experiment + Extrapolate to future implications

Structure:

1. Hook (Transformation): "I Found An X That MIGHT Never Fail"
2. Discovery: How you found this opportunity
3. Experiment: What you tested
4. Results: Metrics and outcomes
5. Pattern recognition: Why this works long-term
6. Future implications: Where this leads
7. How to capitalize: Action steps for readers

When to Use:

• Case studies
• Personal success stories
• Novel approaches you've tested
• "I did this so you don't have to" content

Example:

Hook: "I Found a Content Agent Pattern That MIGHT Never Fail"

Structure:
• The problem: Content creation is time-intensive and inconsistent
• My hypothesis: Framework-driven AI generation could scale quality
• The experiment: Built ContentGen + integrated 123 proven frameworks
• Week 1-4 results: 7,540 ideas generated, 89% pass quality bar
• Why it works: Combines human-proven patterns with AI scale
• The staying power: Frameworks evolve, but patterns are timeless
• Future implication: Framework libraries become competitive moats
• How to start: Pick 3 frameworks, test for 30 days

CTA: "Want my framework setup? Link below"

Platforms: Blog (detailed case study), LinkedIn (professional narrative), Twitter thread (results-focused)

Framework #4: Teacher + Magician

Formula: Educational breakdown + Surprising shortcut/technique

Structure:

1. Hook (How-To): "9 X You Can Build While STILL Y"
2. Constraint acknowledgment: Time/resource limitation
3. Workflow breakdown: 9 specific implementations
4. Time estimates: Honest time investment per workflow
5. Quick wins: Immediate value from each
6. The magic: One surprising insight that accelerates all 9
7. Templates: Ready-to-use starting points

When to Use:

• Tutorial compilations
• "Here's everything you need" guides
• Productivity/efficiency content
• Resource libraries

Example:

Hook: "9 Personal-OS Workflows You Can Build This Weekend"

Structure:
• The challenge: Full-time job, side project dreams
• Why this works: Micro-workflows compound
• Workflow 1: Daily log analyzer (30 min)
• Workflow 2: Content idea aggregator (45 min)
• Workflow 3: Framework matcher (1 hour)
...through Workflow 9
• The magic: They all use the same base agent structure (teach once, apply 9x)
• Templates: GitHub repo with starter code
• Time investment: 6 hours total, lifetime value

CTA: "Clone the repo and start with #1"

Platforms: YouTube (visual tutorials), Blog (comprehensive guide), Course platform

Framework #5: Experimenter + Teacher

Formula: Build something yourself + Teach the process

Structure:

1. Hook (Build Tutorial): "How to Build X With Zero Y in 2025"
2. Belief challenged: Traditional requirements (coding, budget, etc.)
3. Alternative demonstrated: No-code/low-code path
4. Step-by-step walkthrough: Complete journey
5. Decision points: Where to choose between options
6. Common pitfalls: What to avoid
7. Final result: What you built, working demo

When to Use:

• No-code/low-code tutorials
• Accessibility-focused content
• "Anyone can do this" messaging
• Live builds or walkthroughs

Example:

Hook: "How to Build an AI Content System With ZERO Coding in 2025"

Structure:
• The old way: Python, APIs, deployment pipelines
• The challenge: Build ContentGen equivalent with no code
• Tools used: n8n, Airtable, Google Sheets, Make.com
• Step 1: Data collection (Airtable setup)
• Step 2: Framework library (Google Sheets + formulas)
• Step 3: Automation (n8n workflow)
• Step 4: Output formatting (Make.com → platforms)
• Pitfalls: Where no-code gets tricky, workarounds
• Result: Functional content system, zero Python

CTA: "Download my n8n workflow template"

Platforms: YouTube (screen recording), Blog (screenshots + explanations), Course

Framework #6: Investigator + Fortune Teller

Formula: Research deep-dive + Predict market opportunity

Structure:

1. Hook (Curiosity): "Highest Paying X Skills No One's Talking About"
2. Research methodology: How you gathered data
3. Findings: Specific skills with salary data
4. Market gaps: Why these are undervalued now
5. Trend analysis: Why demand is increasing
6. Prediction: Future market state
7. Learning paths: How to acquire these skills
8. ROI analysis: Time investment vs. earning potential

When to Use:

• Career advice content
• Market opportunity analysis
• "Get ahead of the curve" positioning
• Educational product marketing

Example:

Hook: "Highest Paying AI Skills No One's Talking About (2025)"

Structure:
• Research: Analyzed 1,000 AI job postings, salary surveys
• Skill #1: Framework engineering ($120-180K avg) - trend data
• Skill #2: Agentic workflow design ($140-200K) - growing 300% YoY
• Skill #3: AI context architecture ($130-190K) - emerging field
...
• Why undervalued: New disciplines, no formal training yet
• Trend: Companies realizing prompting ≠ production systems
• Prediction: Dedicated roles in 18-24 months
• Learning path: Build real systems, document frameworks
• ROI: 3-6 months learning → $50K+ salary increase

CTA: "Free learning roadmap in description"

Platforms: LinkedIn (professional), YouTube (data visualization), Blog (comprehensive research)

Framework #7: Contrarian + Teacher

Formula: Challenge norm + Teach alternative approach

Structure:

1. Hook (Contrarian): "Get 'Unwanted' X and Start One Of These Y"
2. Conventional wisdom: What everyone thinks is required
3. Challenge: Actually, "unwanted" resources are goldmines
4. Business models: 7 specific applications
5. Case studies: Real examples of each
6. Implementation: Step-by-step for each model
7. Revenue potential: Realistic projections

When to Use:

• Unconventional business ideas
• Opportunity spotting content
• "Hidden in plain sight" narratives
• Entrepreneurship content

Example:

Hook: "7 Content Sources Everyone Ignores (But Shouldn't)"

Structure:
• Conventional: Need "fresh" ideas, original research
• Reality: Best content repurposes "boring" data sources
• Source #1: Git commit messages → learning insights blog
• Source #2: Personal project logs → social media micro-content
• Source #3: Old Stack Overflow answers → updated tutorials
• Source #4: Browser history → trend analysis content
• Source #5: Email threads → case study material
• Source #6: Slack conversations → behind-the-scenes posts
• Source #7: Error logs → debugging tutorial series
• Implementation: Data extraction → framework application → publishing
• Revenue: Each source = content pillar = email list growth

CTA: "My data extraction templates (link)"

Platforms: Blog (detailed guides), YouTube (screen share demos), Email course

Framework #8: Contrarian + Experimenter

Formula: Challenge mainstream + Demonstrate alternative

Structure:

1. Hook (Contrarian): "Why I'm Using X (And You Should Too)"
2. Mainstream approach: What everyone does
3. The problem: Hidden costs/limitations
4. Your alternative: Contrarian choice
5. Experiment details: How you tested
6. Results comparison: Benchmarks and metrics
7. When to use: Applicability guidelines

When to Use:

• Tool comparisons
• Methodology debates
• "Against the grain" positioning
• Technical deep-dives

Example:

Hook: "Why I'm Using LOCAL AI Models (And You Should Too)"

Structure:
• The trend: Cloud-first AI (GPT-4, Claude via API)
• Hidden costs: $, privacy, latency, vendor lock-in
• My alternative: Local models (Llama, Mistral, etc.)
• 30-day experiment: Migrated Personal-OS agents to local
• Benchmarks: Cost ($200/mo → $0), latency (-40%), privacy (100% local)
• Trade-offs: Setup complexity, model quality gaps
• When to use local: Sensitive data, high volume, offline needs
• When to use cloud: Cutting-edge models, low volume, simplicity

CTA: "Local model setup guide (link)"

Platforms: Technical blog, YouTube (benchmarks), Twitter thread (results)

Layering Frameworks: The Compound Effect

Layer 1: Hook (First 3 Seconds)

Use viral-hook-generator to create attention-grabbing opening.

Options:

• Contrarian: "Most people think X. I found the opposite."
• Benefit-Driven: "How I 10x'd X with Y"
• Transformation: "The Z that saved me 10 hours"
• How-To: "I rebuilt X in Y changes (here's how)"

Layer 2: Title/Promise (SEO + Click)

Use youtube-title-optimizer for platform-appropriate title.

Options:

• Build: "Build a FULL X With Y With Z Input!"
• Transformation: "From A to B in C time using D"
• Number: "N patterns that outcome"
• Curiosity: "The X nobody talks about"

Layer 3: Structure (Content Delivery)

Select AI Video Framework based on content type.

Decision Tree:

• Challenging assumption + have data? → Contrarian + Investigator
• Trend analysis + prediction? → Contrarian + Fortune Teller
• Personal case study? → Experimenter + Fortune Teller
• Tutorial compilation? → Teacher + Magician
• Build walkthrough? → Experimenter + Teacher
• Market analysis? → Investigator + Fortune Teller
• Business opportunities? → Contrarian + Teacher
• Tool comparison? → Contrarian + Experimenter

Layer 4: Voice/Tone (Platform Fit)

Use platform-voice-adapter to match audience.

Options:

• Twitter: Concise, punchy, thread-friendly
• LinkedIn: Professional, storytelling, data-driven
• YouTube: Conversational, visual, paced
• Blog: Comprehensive, structured, SEO-optimized

Full Example: Layered Framework

Content: Building an AI agent that generates content using frameworks

Layer 1 - Hook (Contrarian): "Most people think AI content is generic. I built an agent using proven frameworks and it outperforms humans."

Layer 2 - Title (Build Pattern): "Build a FULL Content Generation Agent With Claude Code With 3 SCREENSHOTS!"

Layer 3 - Structure (Experimenter + Fortune Teller):

• The problem: AI content is hit-or-miss quality
• My hypothesis: Human-proven frameworks + AI scale = consistent quality
• The build: ContentGen agent with 123 frameworks
• Results: 7,540 ideas, 89% quality rate
• Why it works long-term: Frameworks evolve slowly, patterns are timeless
• Future implication: Framework libraries become competitive advantages
• How to replicate: Step-by-step agent build

Layer 4 - Voice (Platform-Specific):

YouTube version: "What's up! Today I'm showing you how I built a content agent that's honestly better than me at coming up with viral hooks. We're using Claude Code, 123 proven frameworks from top creators, and literally 3 screenshots to make this happen. By the end, you'll have a working agent that generates hundreds of content ideas. Let's dive in!"

LinkedIn version: "I spent the last month building something that challenges conventional wisdom about AI-generated content.

The result: An autonomous agent that produces content ideas with an 89% quality approval rate—higher than my manual brainstorming sessions.

The secret? Instead of letting AI 'be creative,' I fed it 123 proven frameworks from top creators (Kallaway hooks, YouTube title patterns, etc.) and let it pattern-match.

Here's the complete architecture and what I learned about framework-driven AI:"

Twitter Thread: "Just proved AI content can outperform human creativity (with the right setup) 🧵

1/ The problem: AI content feels generic because it has no framework

2/ My experiment: Built an agent with 123 proven hooks/patterns from viral creators

3/ Results after 30 days: • 7,540 content ideas generated • 89% pass manual quality check • 3x faster than manual ideation

4/ The insight: AI doesn't need to be creative. It needs proven patterns to remix

5/ Full build tutorial (with Claude Code screenshots):"

Quality Checklist

Before publishing framework-mixed content:

Hook Layer:

• Grabs attention in first 3 seconds
• Uses power words from selected framework
• Creates curiosity gap or promise

Title Layer:

• Matches content to proven pattern
• Includes specific numbers/metrics
• Tool/keyword for SEO

Structure Layer:

• Selected framework fits content type
• All sections included (hook → setup → delivery → CTA)
• Flows logically through framework structure

Voice Layer:

• Tone matches platform
• Length appropriate
• Format (bullets, paragraphs, etc.) fits

Overall:

• Each layer amplifies the others
• No contradictions between layers
• Single clear CTA
• Delivers on promise made in hook/title

Integration with Content Workflow

Recommended Stack:

1. Content idea → Social Media Content Agent
2. Hook → viral-hook-generator
3. Title → youtube-title-optimizer
4. Structure → This skill (framework-content-mixer)
5. Voice → platform-voice-adapter
6. Platform-specific polish → linkedin-thought-leader, twitter-thread-builder

Workflow:

Raw content idea
    ↓
Generate hook options (viral-hook-generator)
    ↓
Generate title options (youtube-title-optimizer)
    ↓
Select AI Video Framework based on content type
    ↓
Structure content following framework
    ↓
Adapt voice for platform (platform-voice-adapter)
    ↓
Add platform-specific elements (CTAs, formatting)
    ↓
Ready to publish

Reference Files

See /references/ for:

• ai_video_frameworks_full.json - All 8 frameworks with detailed structures
• framework_combination_guide.md - When to mix which frameworks
• layering_examples.md - 20+ real examples of layered content
• platform_framework_matrix.md - Best framework combinations per platform