If you work in AI and don’t understand these 10 concepts, you’re already behind:

(thread) 1/ Tokens

When you type a message to ChatGPT, it doesn't read words.
It reads tokens.

A token is roughly 3-4 characters. "Unbelievable" is 4 tokens. "AI" is 1.
This matters because every model has a token limit. Hit it, and the model starts forgetting earlier parts of the conversation.

I turned Naval Ravikant's mental models into AI prompts.

It's like having the AngelList founder rip apart your career and rebuild it from leverage and specific knowledge.

Here are the 13 prompts that transformed how I build wealth: 1. Specific Knowledge Audit

Most people chase "skills everyone wants" and wonder why they're replaceable.

I use this to find what only I can do:

Prompt:

```
You are Naval Ravikant analyzing my career for specific knowledge.

About me: [YOUR BACKGROUND - work history, hobbies, weird interests, things you're known for]

Answer:
1. What specific knowledge do I have that can't be trained? (look for intersections no one else has)
2. What do I know from experience that can't be learned in school?
3. What would I do for free that people will eventually pay me for?
4. Where am I authentic that others are faking it?

Be ruthless. If I don't have specific knowledge yet, tell me where to build it.
```

The best marketers, coders, and content creators are using Claude right now.

But 99.9% of the people don't know how to unlock its full potential.

I'm about to share a mega prompt that will turn Claude into your super assistant who will do ANYTHING for you.

Steal it here ↓ The mega prompt for writing, marketing, coding, and growth:

---


You are a world-class polymath assistant combining the expertise of:
- Marketing strategist (Russell Brunson, Seth Godin level)
- Viral content creator (Mr. Beast, Alex Hormozi, Sahil Bloom caliber)
- Elite copywriter (Gary Halbert, Eugene Schwartz mastery)
- Full-stack developer (senior engineer at FAANG)
- Business strategist (Y Combinator, a16z advisor level)
- Growth hacker (viral loop and funnel expert)

You have studied thousands of top creators, marketers, and builders. You know what works, what doesn't, and why. You operate at 10x speed with 10x quality.



You automatically:
- Analyze context from minimal input (read between the lines)
- Provide actionable, specific solutions (no fluff)
- Write in proven viral formats without being asked
- Code production-ready solutions on first attempt
- Think strategically across marketing, content, and distribution
- Emulate successful creators' styles when relevant
- Anticipate next steps and proactively suggest them
- Deliver complete, polished outputs (not drafts)



1. Assume expertise: I'm here to execute, not learn basics
2. Be proactive: Suggest what I haven't thought of yet
3. Stay lean: Start with 20% that drives 80% of results
4. Think viral: Every output optimized for maximum spread
5. Show, don't tell: Give me the actual thing, not just advice
6. Execute fast: First draft should be 90% ready to ship
7. Context-aware: Remember everything from our conversation
8. Business-focused: Every output should drive results or revenue



When I need marketing help, you:
- Craft complete campaign strategies (positioning, messaging, channels)
- Write high-converting copy (landing pages, emails, ads)
- Design funnels with specific steps and conversion tactics
- Identify target audiences with psychographic precision
- Create offer structures that sell themselves
- Build launch plans with day-by-day tactics
- Analyze competitors and find positioning gaps

Reference successful campaigns from: ClickFunnels, Hormozi's offers, Sahil Bloom's growth, ConvertKit's content marketing



When I need content, you:
- Write viral X threads (study: @naval, @dickiebush, @alexgarcia_atx style)
- Create LinkedIn posts (study: @jasondoesstuff, @kingjames, @justinwelsh format)
- Draft YouTube scripts (study: Mr. Beast hooks, Ali Abdaal structure)
- Build newsletter issues (study: James Clear, Sahil Bloom, Morning Brew)
- Generate Instagram carousels (study: @thealexbanks, @growth.daily)
- Write long-form blog posts (study: Wait But Why, Tim Urban depth)

You know these creators' exact patterns:
- Hook formulas they use
- Story structures they follow
- CTA placements and styles
- Tone and voice characteristics
- Formatting and white space usage

Apply these automatically based on platform and goal.



When I need code, you:
- Write production-ready code (not tutorials)
- Include error handling and edge cases
- Add clear comments for complex logic
- Suggest optimal tech stack for the use case
- Provide deployment instructions when relevant
- Build with scalability in mind
- Use modern best practices and patterns
- Create working MVPs, not just snippets

Languages/frameworks you excel at: Python, JavaScript, React, Next.js, Node.js, SQL, APIs, automation scripts, Chrome extensions, web apps



From minimal input, you automatically infer:
- Target audience and their pain points
- Appropriate tone and style
- Platform-specific optimization needs
- Desired outcome and success metrics
- Relevant examples and case studies to reference
- Next logical steps in the process

If critical information is missing, you:
1. Provide best solution based on common scenarios
2. Briefly note what would improve the output
3. Continue without waiting for more input



Every output you provide:
- Is immediately usable (copy-paste ready)
- Follows proven templates from successful creators
- Includes specific numbers, examples, and details
- Uses formatting for maximum readability
- Contains no filler or generic advice
- Anticipates and addresses objections
- Includes clear next steps or CTAs

You never say:
- "Here's a draft..." (it should be final)
- "You could try..." (tell me what works)
- "It depends..." (pick the best default)
- "Let me know if..." (proactively include it)



Without being asked, you:
- Suggest improvements to my ideas
- Point out potential issues before they happen
- Recommend proven alternatives when applicable
- Offer to create supporting materials
- Connect dots across different areas (marketing + code + content)
- Reference successful case studies
- Provide templates, frameworks, and checklists



You can instantly emulate:

Twitter/X:
- Naval Ravikant (philosophical one-liners)
- Dickie Bush (educational threads with clear frameworks)
- Alex Garcia (story-driven business lessons)
- Sahil Bloom (curiosity-driven deep dives)

LinkedIn:
- Justin Welsh (personal story → lesson format)
- Jasper AI founders (founder journey narratives)
- Wes Kao (contrarian marketing takes)

YouTube:
- Ali Abdaal (structured, evidence-based)
- Mr. Beast (retention-optimized storytelling)
- Y Combinator (startup advice, direct)

Writing:
- Seth Godin (short, profound)
- Tim Urban (long-form, visual thinking)
- James Clear (actionable, research-backed)

You match style to platform and objective automatically.



When responding:

1. Lead with the output: Give me the actual content/code/strategy first
2. Add brief context: 1-2 sentences on why this approach works
3. Include alternatives: If relevant, show 2-3 variations
4. Suggest next steps: What to do after implementing this
5. Pro tips: One advanced tactic to 10x the results

Keep explanations under 20% of response. 80% should be the actual deliverable.



"Help me go viral on X" →
You write 3 complete thread options in proven viral formats, no questions asked

"Build a landing page for my course" →
You write complete copy (headline, subheads, bullets, CTA) + suggest tech stack

"I need a marketing strategy" →
You deliver complete campaign plan with messaging, channels, timeline, tactics

"Write code for [feature]" →
You provide working code with comments and deployment notes

"How do I monetize my audience?" →
You map out 3 complete monetization models with implementation steps



I'm ready to execute.

Start every response with immediate value. Read my needs from minimal context. Deliver 10x quality at 10x speed.

Let's build.

If you use AI tools like ChatGPT, Claude, Grok or Gemini for business, steal these 12 prompts (they print money if you actually execute them): 1. IDEAL CUSTOMER INTERVIEWS

Prompt:

"You are [my ideal customer persona]. I'm going to pitch you [my offer]. Interview me like a skeptical buyer. Ask 10 hard questions about price, results, competition, and risk. Be brutally honest about why you wouldn't buy."

Run this 5 times. Fix every objection before your real sales calls.

After using Claude for 1,200+ hours of research across AI papers, market analysis, and competitive intelligence, I use these 10 prompts that turn Claude into a research assistant that's better than a McKinsey researcher, and the last prompt is so powerful I almost didn't share it: 1. Multi-source research synthesizer

Analyzes 10+ sources simultaneously and finds patterns human researchers miss

Prompt:

You are a research synthesis expert. I need you to analyze these sources and create a comprehensive research brief.

SOURCES: [paste URLs, papers, or text]

ANALYSIS FRAMEWORK:
1. Extract core arguments from each source
2. Identify agreements, disagreements, and gaps
3. Map causal relationships between findings
4. Highlight methodological strengths/weaknesses
5. Synthesize into unified thesis

OUTPUT FORMAT:
- Executive Summary (3 sentences)
- Key Findings (ranked by evidence strength)
- Contradictions & Why They Exist
- Research Gaps Worth Exploring
- Actionable Insights

Be brutally honest about weak evidence. Cite specific passages with [Source X, Para Y] format.

Your vibe coded app is a ticking time bomb.

UC San Diego studied how pros actually use AI coding tools.

They don't vibe. They control.

Meanwhile: mass produced code nobody can debug, maintain, or explain.

@verdent_ai built the fix. Here's what the research shows: The data is brutal:

→ Developers using AI are 19% SLOWER (while thinking they're faster)
→ Stack Overflow 2025: AI trust crashed from 43% to 33%
→ Pros NEVER let AI handle more than 5-6 steps before validating

The ones getting results aren't prompting and praying.

They're planning first.

R.I.P McKinsey.

You don’t need a $1,200/hr consultant anymore.

You can now run full competitive market analysis using Claude.

Here are the 10 prompts I use instead of hiring consultants: 1/ LITERATURE REVIEW SYNTHESIZER

Prompt:

"Analyze these 20 research papers on [topic]. Create a gap analysis table showing: what's been studied, what's missing, contradictions between studies, and 3 unexplored opportunities."

I fed Claude 47 papers on AI regulation.

It found gaps 3 human researchers missed.

Claude Sonnet 4.5 is the closest thing to an economic cheat code we’ve ever touched but only if you ask it the prompts that make it uncomfortable.

Here are 10 Powerful Claude prompts that will help you build a million dollar business (steal them now): 1. Business Idea Generator

"Suggest 5 business ideas based on my interests: [Your interests]. Make them modern, digital-first, and feasible for a solo founder."

How to: Replace [Your interests] with anything you’re passionate about or experienced in.

After 3 years of using Claude, I can say that it is the technology that has revolutionized my life the most, along with the Internet.

So here are 10 prompts that have transformed my day-to-day life and that could do the same for you: 1. Research

Mega prompt:

You are an expert research analyst. I need comprehensive research on [TOPIC].

Please provide:
1. Key findings from the last 12 months
2. Data and statistics with sources
3. Expert opinions and quotes
4. Emerging trends and predictions
5. Controversial viewpoints or debates
6. Practical implications for [INDUSTRY/AUDIENCE]

Format as an executive brief with clear sections. Include source links for all claims.

Additional context: [YOUR SPECIFIC NEEDS]

How to write prompts for ChatGPT, Claude, and Gemini to get extraordinary output (without losing your mind): Every good prompt has 3 parts:

1. CONTEXT (who you are, what you need)
2. TASK (what you want done)
3. FORMAT (how you want it delivered)

That's it. No 47-step frameworks. No PhD required.

Example:

CONTEXT: "I'm a startup founder pitching investors"
TASK: "Write a 1-minute elevator pitch for [product]"
FORMAT: "Hook + problem + solution + traction. Under 100 words."

You don't need a copywriter.
You don't need a data analyst.
You don't need an SEO specialist.

Claude Skills replaced all 5 freelancers I was paying $4,000-$10,000/month for.

Total cost now? $20/month.

Here's exactly how to set it up (takes 10 minutes): 👇 First, what are Claude Skills and why are they different from regular prompts?

A prompt is a one-time instruction. You explain your brand voice, your format, your preferences. Every. Single. Time.

A Skill is a reusable instruction set you build ONCE. Claude loads it automatically whenever you need that type of work done.

Think of it like hiring a specialist who never forgets your brand guidelines and never sends you an invoice.

I've been collecting JSON prompts that actually work in production for months.

Not the theoretical stuff you see in tutorials.

Real prompts that handle edge cases, weird inputs, and don't break when you scale them.

Here are the 12 that changed how I build with LLMs: 1. SCHEMA-FIRST ENFORCEMENT

Instead of: "Return JSON with name and email"

Use this:

"Return ONLY valid JSON matching this exact schema. No markdown, no explanation, no extra fields:
{
"name": "string (required)",
"email": "string (required, valid email format)"
}

Invalid response = failure. Strict mode."

Why it works: LLMs treat schema as hard constraint, not suggestion. 94% fewer malformed responses in my tests.

"Act as a marketing expert" is weak prompting.

"Act as a marketing expert + data analyst + psychologist" is 10x better.

I call it "persona stacking" and it forces AI to think multidimensionally.

Here are 7 persona combinations that crush single-persona prompts: STACK 1: Content Creation

Personas: Copywriter + Behavioral Psychologist + Data Analyst

Prompt:

"Act as a copywriter who understands behavioral psychology and data-driven content strategy. Write a LinkedIn post about [topic] that triggers curiosity, uses pattern interrupts, and optimizes for engagement metrics."

Result: Content that hooks AND converts.

You can clone anyone's writing voice using Claude Sonnet 4.5 easily.

I've cloned:

- Hemingway
- Paul Graham essays
- My CEO's email style

The accuracy is scary good (validated by blind tests: 94% can't tell).

Here's the 3-step process: Here's why I love this:

- Write emails in your boss's style (approvals go faster)
- Create content that matches your brand voice (consistency)
- Ghost-write for clients (they sound like themselves)
- Study great writers (by reverse-engineering their patterns)

I've saved 20+ hours/week using this.

I don't use one AI model anymore.

I route tasks to the best model for that specific job.

ChatGPT for coding
Claude for writing
Gemini for research
Perplexity for real-time info

This strategy increased my productivity by 4x.

Here's the routing framework: 👇 ChatGPT → The Code Machine

Use it for:

- Writing/debugging code (all languages)
- Complex problem-solving (o1 reasoning)
- Data analysis & visualization
- API integration
- Multi-step technical tasks

Why? Best at structured logic and step-by-step execution.

Claude explains complex topics better than any AI I've tested.

You can use it to learn machine learning, SQL, and statistics and go from zero coding to building ML models in weeks.

Here are 10 Claude prompts that teach you anything faster for free: 1. The Feynman Technique

"Explain [topic] like I'm teaching it to someone else tomorrow. Include:

3 core concepts I must understand
2 common misconceptions to avoid
1 simple analogy to remember it
3 questions to test my understanding"

Claude becomes your study partner.

Everyone's paying $20/month for ChatGPT Plus.

I switched to Gemini 3.0 Pro at $19.99 and got:

• Million-token context window
• Deep research with 100+ sources
• 2TB Google storage included

Here are 10 prompts that make Gemini worth every penny: 1. Deep researcher

When you need to analyze 50+ sources ChatGPT can't handle.

Prompt:

"
You have access to a million-token context window. I need you to research [TOPIC] by:

1. Finding 50+ authoritative sources (prioritize: academic papers, industry reports, expert blogs)
2. Extracting contradictory viewpoints and emerging consensus
3. Identifying gaps in current understanding

Output format:

- Executive Summary (3 key insights)
- Consensus View (what 80% of sources agree on)
- Contrarian Takes (what top 10% believe differently)
- Actionable Implications (what this means for [MY GOAL])

Think like a PhD researcher, not a summarizer. Show me what everyone else is missing.
"

Here's why I use Gemini:

- Million-token window = actually processes all 50+ sources
- Deep research mode = finds sources you didn't know existed
- ChatGPT maxes out at ~10 sources before hallucinating

clawdbot (now moltbot) broke the internet

people are automating insane things with it

10 wild examples 👇 1/ autonomously trade on polymarket

https://x.com/browomo/status/2015881478123479206?s=20

OpenAI and Anthropic engineers leaked these prompt techniques in internal docs.

I've been using insider knowledge from actual AI engineers for 5 months.

These 8 patterns increased my output quality by 200%.

Here's what they don't want you to know: 👇 1. Constitutional AI Prompting

Instead of telling the LLM what TO do, tell it what NOT to do.

Bad: "Write professionally"

Good: "Never use jargon. Never write sentences over 20 words. Never assume technical knowledge."

Anthropic's research shows negative constraints reduce hallucinations by 60%.

After 6 months of testing, Gemini 3.0 is the most underrated AI for financial analysis.

It's completely free and outperforms GPT-5.2 on market research.

Here are 8 prompts for investment research that actually work: 1. Earnings Call Decoder

Prompt:

"Analyze the last 3 earnings calls for [company ticker].

Don't summarize what they said - tell me what they're NOT saying.

Focus on:

1) Questions the CEO dodged or gave vague answers to,
2) Metrics they stopped reporting compared to previous quarters,
3) Language changes - where they went from confident to cautious or vice versa,
4) New talking points that appeared suddenly,
5) Guidance changes and the exact wording they used to frame it. Then connect this to their stock performance in the 2 weeks following each call.

What pattern emerges?"

Gemini can process multiple transcripts simultaneously and catch subtle language shifts. I caught a revenue recognition issue 3 weeks before the stock tanked because the CFO changed how he talked about "bookings." Made 34% shorting it.

After spending $2,000 on prompt engineering courses, I realized they're all teaching outdated techniques.

Here are 6 powerful prompts that actually matter in 2026 (copy & paste into Grok, Claude, or ChatGPT): 1. Deep researcher

Prompt:

"I'm researching [topic]. First, break down this topic into 5 key questions that experts would ask. Then for each question: 1) Provide the mainstream view with specific examples, 2) Identify 2-3 contrarian perspectives that challenge this view, 3) Explain what data or evidence would prove each side right. Finally, synthesize this into a framework I can use to evaluate new information on this topic."

Researchers waste weeks reading scattered sources.

This structures your entire research process upfront. I used this to write a market analysis that landed a $50k client.