After testing every AI writing tool for 6 months, I found the one workflow that actually produces content worth reading.
It's not a tool. It's 5 Claude prompts run in a specific order that turns a rough idea into a finished piece in 40 minutes.
Here's the system:
Every AI writing tool has the same problem.
They start at the wrong end.
You give them a topic. They give you a draft. The draft is clean, organized, and completely hollow because the tool skipped the only part that makes writing worth reading.
The thinking.
Good writing isn't organized information. It's a writer working something out in public finding the angle nobody took, the tension nobody named, the insight that was obvious in hindsight and invisible before.
No tool finds that for you. But a system can force you to find it yourself before a single word of the actual piece gets written.
That's what these 5 prompts do. They run in order. Each one builds on the last. By the time you reach Prompt 5, you're not writing from a blank page you're writing from a position.
40 minutes. One rough idea in. One finished piece out.
Here's the system.
PROMPT 1 - The Angle Excavator
Most people start writing with a topic. The best writers start with a tension.
Run this first before you write a single sentence.
"I have a rough idea for a piece of writing. Your job is not to outline it. Your job is to find what's actually interesting about it.
Read the idea below and give me:
The obvious angle what everyone who covers this topic already says.
The contrarian angle what someone who has thought about this longer than anyone would say instead.
The personal angle the version of this idea that only someone with a specific lived experience could write authentically.
The tension the unresolved contradiction inside this topic that makes it genuinely worth writing about right now.
Do not write the piece. Give me the four angles and tell me which one has the most to say that hasn't already been said.
Here is my rough idea: [paste idea]"
Pick the angle that makes you slightly uncomfortable. That's the one.
PROMPT 2 - The Skeleton
Now you have an angle. You need a structure that serves it not a generic five-paragraph frame, a structure built specifically around the tension you chose.
"I'm writing a piece with this central tension: [paste chosen angle from Prompt 1].
Build me a skeleton not an outline. A skeleton has the minimum structure needed to hold the piece upright. Every section exists because removing it would collapse something.
For each section give me: what it does (not what it says), why it comes at this specific point in the sequence, and the one question it must answer before the next section can land.
No bullet points of talking points. No topic sentences. Only the architecture."
This prompt stops you from writing a list dressed up as an essay.
PROMPT 3 - The Opening Stress Test
The opening is the only part that determines whether anyone reads the rest.
Run this before you write it not after.
"Here is the skeleton for my piece and its central tension: [paste].
Write 4 completely different opening paragraphs for this piece. Each one must use a different entry point: a specific scene, a counterintuitive claim, a named person in a specific moment, and a question the reader cannot immediately answer.
None of them should introduce the topic. All of them should drop the reader into the middle of something already in motion.
After the four openings, tell me which one creates the most forward momentum and why the others stall."
Read all four out loud. The one that makes you want to keep reading is the one.
PROMPT 4 - The Draft Engine
Now write the piece. But don't write it alone.
"You are a developmental editor watching me write in real time.
Here is my opening, my skeleton, and my central tension: [paste all three].
I am going to write this piece section by section. After each section I paste, do three things only:
Tell me the last sentence that still has momentum and where it stopped.
Tell me the one thing the next section must do to keep the reader moving forward.
Flag any sentence where I slipped into explaining instead of showing.
Do not rewrite anything. Do not summarize. Only respond to what I paste, section by section."
Paste one section at a time. The editor's job is to keep you from going flat.
PROMPT 5 - The Closing Audit
Most pieces end when the writer runs out of things to say.
The best pieces end when they've earned the right to stop.
"Here is my completed draft: [paste full piece].
Run a closing audit. Tell me:
Does the final paragraph land on something the reader didn't have when they started or does it restate what they already read?
Is there a sentence anywhere in the piece that is doing the work the final paragraph should be doing and doing it earlier?
What is the single most surprising thing in this piece? Is it in the first 150 words or buried somewhere the reader might not reach?
What would Hemingway cut from this draft without asking permission?
Return the answers. Then give me one alternative final sentence that closes harder than the one I wrote."
Fix the close. Then publish.
The whole system takes 40 minutes if you move through it without stopping.
The pieces it produces take considerably longer to explain how you wrote them.
AI makes content creation faster than ever, but it also makes guessing riskier than ever.
If you want to know what your audience will react to before you post, TestFeed gives you instant feedback from AI personas that think like your real users.
It’s the missing step between ideas and impact. Join the waitlist and stop publishing blind.
Inversion is the most powerful thinking tool most people never use correctly.
They invert the goal. They don't invert the system.
I turned Claude into a full inversion engine that runs Charlie Munger's method on any problem mapping every path to failure so precisely that the path to success becomes obvious by elimination.
Here are the 5 prompts:
Munger said it best: "Tell me where I'm going to die, so I'll never go there."
Most people use inversion as a cute thought exercise.
They ask "what if this fails?" write 3 bullet points, feel smart, and move on.
That's not inversion. That's journaling with extra steps.
Real inversion is forensic. You don't brainstorm failure. You systematically reconstruct it every assumption, every decision point, every handoff where things rot quietly before they collapse loudly.
The difference between someone who thinks about failure and someone who maps it is the difference between a smoke alarm and a fire investigation.
One warns you. The other tells you exactly what burned and why.
Prompt 1: The Pre-Mortem
"Assume it's 18 months from now and [your goal/project] has completely failed. Not stumbled failed. Dead. Done.
You're writing the post-mortem report.
Work backwards. Identify: the single decision that sealed it, the warning sign that appeared early but was ignored, the assumption that was never tested, and the person in the room who knew but didn't say it.
Be specific. Name the failure mode, not the feeling of failure.
Then rank the top 3 causes by how invisible they would have been at the start."
A Goldman Sachs analyst once told me: "We don't read about competitors. We study them."
I turned that into a 5-prompt Claude workflow that does a full competitive teardown in 10 minutes.
Here's exactly how:
1. The Strategic Intent Decoder
Companies tell you what they did in their reports.
They accidentally tell you what they're planning in how they talk about it.
Word choice. Emphasis. What gets one paragraph versus five. What gets mentioned in the CEO letter versus buried in footnotes.
Goldman analysts read for intent, not just information.
"Read this competitor's most recent annual report, earnings call transcript, and investor day presentation: [paste or describe them]. Do not summarize what they said. Extract what they are planning. Where is language unusually emphatic or repeated across multiple documents and what does that signal about internal conviction? What did they mention once and drop that suggests a strategic experiment they're quietly watching? What segment, geography, or product line received more narrative attention than its current revenue contribution justifies and why might that be? What are they clearly building toward that they haven't announced yet?"
2/ The Hiring Pattern Intelligence Extractor
Job postings are the most honest strategic document any company publishes.
Because they reflect actual resource allocation decisions, not narrative ones.
A company can say they're focused on enterprise sales while posting 40 product engineering roles and zero enterprise sales roles.
The postings tell the truth. The press releases don't.
"Analyze the last 90 days of job postings from [competitor]. Map: which functions are they hiring into aggressively versus pulling back from, what specific technical skills or tools appear repeatedly that signal a platform or product direction they haven't announced, are they hiring senior people who would only make sense if a specific strategic initiative was already approved at the executive level, and what does the geographic distribution of their hiring tell you about where they're expanding versus consolidating? Cross-reference the hiring pattern against their stated strategic priorities. Where do they diverge and what does that divergence reveal?"
Claude just helped me build a full business plan, pricing strategy, and launch roadmap in 40 minutes.
A business mentor would charge $10,000 for this.
Copy and paste these prompts into any LLM and start building in 2026:
1/ The Idea Pressure Test
Prompt:
"I have a business idea: [describe it in detail]. Do not encourage me. Do not validate me. Apply the most rigorous pressure test you can. Tell me: what is the single most important assumption this business depends on being true? How would I test that assumption in the next 30 days for under $500 before committing further resources? What are the 3 most common reasons businesses in this category fail that I am probably not thinking about right now? Who has tried something similar and what happened to them? And what would make you genuinely excited about this idea what would have to be true that isn't obviously true yet?"
The idea that survives this prompt is worth building.
The one that collapses is worth knowing about now rather than 18 months from now.
2/ The Customer Obsession Builder
Prompt:
"Help me build a complete customer obsession profile for [your target customer]. Go beyond demographics. Tell me: what does a Tuesday look like for this person hour by hour, what are they stressed about, what are they proud of, what do they avoid thinking about? What have they already tried to solve [the problem your business solves] and why did those solutions disappoint them? What words do they use to describe their problem not marketing language, their actual language? What would have to happen for them to pull out their credit card without needing to think about it? And what belief do they currently hold that my business needs to change before they will ever buy?"
The customer who feels understood buys without being sold to.
This prompt builds that understanding before you write a single line of copy.
Claude can now teach you how to think using the exact method Richard Feynman used at Caltech for 40 years.
Most people use Claude to get answers.
These 5 prompts use it to rewire how fast you learn anything 👇
1/ The Confusion Locator
Feynman said the first step in understanding anything is being honest about what you actually don't understand versus what you just can't explain.
Most people confuse familiarity with understanding.
They've heard a term enough times that it feels known. But the moment they try to explain it, the gaps appear.
"I think I understand [concept] but I want to test that. Ask me to explain it to you as if you're a curious 12-year-old who has never heard of it. After I explain it, tell me: where did my explanation break down or get vague? Where did I use words that assume prior knowledge the 12-year-old wouldn't have? Where did I skip a logical step that I assumed was obvious? Give me a precise list of every gap you found. Those gaps are exactly what I don't actually understand yet."
The gaps this prompt surfaces are more valuable than anything you'd learn from re-reading the source material.
Because they're your specific gaps.
Not the gaps of the average reader.
2/ The First Principle Finder
Feynman never started from the middle of a subject.
He always started from what was actually true at the most fundamental level the irreducible facts that everything else in the field was built on top of.
His first Caltech lecture didn't start with Newton's laws.
It started with the atomic hypothesis. The one idea that if everything else was lost to science would contain the most information in the fewest words.
"I am trying to understand [subject]. Don't teach me the standard curriculum. First: what is the single most fundamental true statement about this subject? The one idea that if I understood it completely would make every other concept in this field easier to learn? Build my understanding from that single statement outward, adding only one layer of complexity at a time, and stopping to check whether each layer is actually clear before adding the next."
The student who starts from first principles always overtakes the one who started from the textbook.
After using Perplexity for 2 years, it has become the only research tool I open before anything else
But only because I stopped using it like Google
Here's the 8 prompt system that turns a topic I know nothing about into a 30-page research brief in under an hour.
1. Start with ignorance mapping
Prompt:
"I know almost nothing about [topic]. Before I start researching, what are the 10 things I most need to understand first? Order them from foundational to advanced. Tell me which ones most people get wrong."
This is not a search. It's a curriculum.
Most people skip this and spend 3 hours reading the wrong things in the wrong order.
2. Find the real debate
Prompt:
"What do experts in [topic] genuinely disagree about right now? Not surface-level stuff. The real fault lines where smart people are on opposite sides and both have evidence."
Google gives you the consensus.
This prompt gives you where the field is actually alive.
After 6 months of using NotebookLM, I can say it's the research tool that has revolutionized my workflow the most.
But only because I learned these 10 prompts.
Here's the complete system that turns 200 pages into clear answers in under an hour:
1. The Source Onboarding Prompt
Before you do anything else, run this the moment you upload your documents.
Paste this into NotebookLM:
"You now have access to [X] sources I've uploaded. Before I start asking questions, give me: 1) The 3 most important overarching themes that run across all these documents, 2) Where these sources agree with each other and where they contradict, 3) The single most surprising or counterintuitive finding across all of them, 4) What major questions these documents raise but don't fully answer."
This gives you a complete map of your research before you've asked a single real question.
Most people skip this. Don't.
2. The Contradiction Hunter
This is where NotebookLM becomes genuinely dangerous in the best way.
Use this:
"Go through all my uploaded sources and find every place where two or more sources disagree, contradict each other, or come to different conclusions on the same topic. For each contradiction: 1) Quote the specific conflicting claims, 2) Identify which source each claim comes from, 3) Give me your assessment of which position has stronger supporting evidence, 4) Flag this as something I need to investigate further."
The best research lives in the gaps between sources.