PromptSharpPrompt LibraryLearning › Three-pass reading of a hard text — without outsourcing the thinking

Dense ReadingFREE

Three-pass reading of a hard text — without outsourcing the thinking

A dense paper or chapter you need to actually understand, not skim. Use the model to scaffold your reading, not replace it.

The prompt — copy and run it

You are helping me READ a hard text properly. You must not read it for me — you scaffold, I read. I will paste the text (or a section of it).

PASS 1 — MAP (before I read): give me the structure — what claim is this making, what's the shape of the argument, what are the 5 terms I must understand first (define each in one plain sentence), and what should I be looking for as I read.

PASS 2 — CHECK (after I read, I'll tell you what I think it said): compare my summary to the text. Tell me what I got right, what I misread, and — most importantly — what I skipped over that matters. Quote the exact lines I missed.

PASS 3 — INTERROGATE: now ask ME the four questions that separate reading from understanding: What is the strongest objection to this? What evidence would change the conclusion? What does the author assume without arguing for it? Where would this fail in practice?

Then: THE ONE PARAGRAPH — after my answers, write the single paragraph I should remember in six months, and mark anything in it that is YOUR inference rather than the text's claim.

Inputs: [THE TEXT] · [WHY I'M READING IT] · [MY BACKGROUND IN THIS AREA]

Rules: Quote the text rather than paraphrasing when accuracy matters. Never assert something the text does not say — clearly separate 'the text claims' from 'I infer'. If the text is beyond what you can verify, say which parts you're unsure about. Do not summarize Pass 2 before I've given you my own summary — that defeats the entire purpose. Do not invent facts, sources, citations, links, or specifics you cannot support — say "I'm not sure" instead. Never paste confidential, client, medical, or personally identifying information into a consumer AI tool, and don't ask me for any.

How to use this prompt

  1. Copy the full prompt above with the Copy button.
  2. Fill in your inputs. Replace each bracketed placeholder with your specifics: [THE TEXT][WHY I'M READING IT][MY BACKGROUND IN THIS AREA]
  3. Paste into ChatGPT, Claude, or Gemini and run. Read the reality guardrail below before you act on the output.

Why this prompt works

Summarize-this destroys the learning and leaves you fluent-sounding and empty. The three-pass structure keeps YOU doing the reading while the model does what it's good at: pre-loading vocabulary, catching what you skipped, and cross-examining your understanding. The infer-vs-claim separation is the anti-hallucination seam.

Get a prompt like this every day

The PromptSharp Learning Brief ships one desk-ready prompt every weekday — free on the web today. Free forever. Today's Learning issue is live on the web right now — subscribe and we email you the sample issue immediately, then the Learning daily every weekday as its email edition ships. Unsubscribe anytime.

Subscribe free → Read a sample issue
Reality guardrail: this prompt makes the model reason from data you paste — it does not source or verify facts for you. Check every claim, keep confidential data out of consumer AI tools, and follow your employer's AI-use policy.

Frequently asked

When should I use this prompt?

A dense paper or chapter you need to actually understand, not skim. Use the model to scaffold your reading, not replace it.

Why does this prompt work?

Summarize-this destroys the learning and leaves you fluent-sounding and empty. The three-pass structure keeps YOU doing the reading while the model does what it's good at: pre-loading vocabulary, catching what you skipped, and cross-examining your understanding. The infer-vs-claim separation is the anti-hallucination seam.

What mistake does this prompt help you avoid?

The one-click 'summarize this paper' habit produces confident-sounding readers who cannot defend a single claim. Withholding the model's summary until after yours is the structural fix.

Related Learning prompts

Learning

Roadmap that ends in a thing you can show, not a course you finished

You want to learn a hard skill and the internet has given you 40 resources and no order. Get a path that terminates in evidence.…

Learning

Practice project at the edge of your ability — with the answers withheld

You've followed the tutorials. Now get a project that's genuinely hard for you, with a coach that refuses to just hand you the sol…

Learning

Socratic drill: make it test you instead of telling you

You 'know' the material until someone asks you a question. Turn the model into an examiner that finds the gap.…

Learning

Explain it to a skeptic and let them find the holes

The final check: teach the concept out loud and get cross-examined by someone who won't nod along.…

Students

Turn a dense reading into a study-ready map (in your own words)

You have a 30-page article due tomorrow and no time to reread it three times. Build understanding, not a summary you'll forget.…

Sales

Coverage check: is there really enough pipeline to hit the number?

You're building next quarter's forecast. Pressure-test whether current pipeline actually covers the goal, and quantify the build g…

Law

Redline playbook check: compare a draft against your standard positions

A counterparty sent their paper. Flag where it deviates from your playbook and what to push back on — as a first pass for an attor…

All Learning free prompts

The PromptSharp Learning Brief page — five full free prompts plus today's issue.

PromptSharp Daily — free

The cross-vertical sampler: one sharp, copy-paste prompt each day, rotating across the roster. Two things in one brief: you get better at AI and prompting, and you see the sharpest prompts from across the network.

Double-opt-in. Unsubscribe anytime. No spam, ever.

Better together
Make prompts remember you: Brainfile

Even a sharp prompt starts from zero unless your AI knows you. Brainfile is persistent context — your work, voice, and priorities loaded into every session. Brainfile is the memory; PromptSharp is the playbook. Together they compound — the same prompt gets sharper because it runs on YOUR context.

Set up your brainfile →

Want both? The All-Access + Brainfile annual bundle covers the pair.

PromptSharp prompts are drafted with AI assistance and human-reviewed. They structure how a model reasons over data you provide — they do not source or verify facts for you, and you own every output. Nothing here is financial, legal, tax, or investment advice. Never paste confidential, client, or material non-public information into consumer AI tools; follow your employer's AI-use policy. © 2026 PromptSharp.