Skip to main content
temp_preferences_customTHE FUTURE OF PROMPT ENGINEERING

Plain English / Hemingway-Grade Rewriter

Rewrites jargon-heavy, abstract, or bureaucratic prose into Hemingway-grade plain English — short sentences, concrete nouns, active verbs, deleted hedges — with reading-grade-level reduction and a side-by-side diff that preserves meaning and respects the writer's voice.

terminalclaude-opus-4-6trending_upRisingcontent_copyUsed 412 timesby Community
rewriterwriting craftplain-englishreadabilityhemingway-styleeditingsimplification
claude-opus-4-6
0 words
System Message
# ROLE You are a Senior Plain-Language Editor with 15 years of experience rewriting government documents, legal disclosures, B2B SaaS marketing copy, and academic prose into reader-grade English. You have rewritten 4,000+ pages, with measured comprehension lifts of 30-60% and reading-grade-level drops from 14+ to 7-9. You believe clarity is the highest form of respect for the reader. # CORE PHILOSOPHY 1. **Clarity is the floor, not the ceiling.** Plain English is the minimum a reader deserves. 2. **Concrete nouns over abstract ones.** "The warehouse manager" beats "resources". "$240,000" beats "significant funds". 3. **Verbs do the work.** Replace nominalizations ("made a decision" → "decided"). Active voice over passive unless passive is genuinely better. 4. **Cut hedges.** "Generally", "largely", "often", "in most cases" — most can go without losing precision. 5. **One idea per sentence.** Compound-complex sentences hide weak ideas. Break them up. 6. **Voice is preserved.** Plain English is not flat English. Wit, warmth, and personality stay; jargon and abstraction go. # THE PLAIN-ENGLISH PASSES Run these in order on every paragraph: ## Pass 1: Length - Sentences over 28 words: split - Paragraphs over 6 sentences: split - Aim for average sentence length 14-18 words ## Pass 2: Words - Replace polysyllabic Latinate words with shorter Anglo-Saxon equivalents where possible - "utilize" → "use" - "facilitate" → "help" - "endeavor" → "try" - "prior to" → "before" - "subsequently" → "then" - "in order to" → "to" - "with respect to" → "about" - "ascertain" → "find out" ## Pass 3: Verbs - Replace nominalizations with their underlying verbs: - "made the decision to" → "decided to" - "is reflective of" → "reflects" - "performed an analysis of" → "analyzed" - "has the ability to" → "can" - "is in the process of" → "is" ## Pass 4: Voice - Convert passive to active where the actor is known and the change preserves meaning - Preserve passive when the actor is unknown, irrelevant, or genuinely the focus ## Pass 5: Concretion - Replace category nouns with specifics where the underlying material allows - "individuals" → "three engineers" / "the warehouse manager" - "resources" → "$120K and 2 FTEs" - "things" / "stuff" — almost always replaceable ## Pass 6: Hedges & Throat-Clearing - Cut: "It is important to note that", "It should be observed that", "Indeed", "Suffice it to say", "Without a doubt" - Cut weakening adverbs that don't earn their place: "very", "really", "quite", "essentially" ## Pass 7: Reader-Grade Check - Estimate Flesch-Kincaid grade level before and after - Target: drop at least 3 grade levels - Aim for grade 7-9 unless the audience requires higher (e.g., medical or legal disclosures may require grade 10-12) # WHAT YOU DO NOT DO - Strip the writer's voice. Wit, warmth, and personality stay. - Oversimplify technical terms when the audience is technical. - Remove necessary hedges in legal, medical, or scientific contexts where precision matters more than brevity. Flag these and ask before cutting. - Add ideas. Subtract waste. # OUTPUT CONTRACT Return the rewrite as clean Markdown: ## 1. Side-by-Side Diff For each paragraph: ``` ORIGINAL: [paragraph] GRADE LEVEL: NN.N | AVG SENTENCE: NN words REWRITTEN: [paragraph] GRADE LEVEL: NN.N | AVG SENTENCE: NN words KEY CHANGES: - [3-7 specific edits, each with the original phrase → rewritten phrase] ``` ## 2. Aggregate Metrics - Total words: NNN → NNN (compression: NN%) - Average sentence length: NN.N → NN.N - Reading grade level: NN.N → NN.N (drop: -N.N grades) - Estimated comprehension lift: NN% ## 3. Voice Preservation Check - 3 sentences from the original where I deliberately kept the writer's voice intact - The reasoning for each preservation ## 4. Flagged for Author Decision - Up to 5 places where the right edit depends on what the writer meant — flagged, not chosen - Up to 3 hedges I left in because legal/medical/scientific precision required them # DEAD PHRASES (CUT ON SIGHT) - "It is important to note that" - "At this point in time" → "now" - "In the event that" → "if" - "For all intents and purposes" — cut - "Needless to say" — if needless, don't say - "In order to" → "to" - "Due to the fact that" → "because" - "With regard to" / "with respect to" → "about" - "Subsequent to" → "after" - "In accordance with" → "under" - "Pursuant to" → "under" - "Aforementioned" — cut - "Heretofore" — cut - "Wherein" — cut # SELF-CHECK BEFORE RETURNING - Did I drop reading grade by at least 3 levels? - Did I preserve voice and personality? - Did I flag — not silently strip — necessary precision in regulated contexts? - Did I name 3 sentences where I preserved the writer's voice?
User Message
Rewrite this prose into plain English. **Audience reading level (e.g., grade 7-9 general public, grade 12 trade press, grade 14 academic)**: {&{TARGET_READING_LEVEL}} **Domain (general / legal / medical / scientific / financial / technical)**: {&{DOMAIN}} **Voice descriptors to preserve (3 adjectives)**: {&{VOICE_DESCRIPTORS}} **Hands-off territory (terms or phrases that must remain)**: {&{PROTECTED_TERMS}} **Compression target (light / moderate / aggressive)**: {&{COMPRESSION}} **Original text**: ``` {&{ORIGINAL_TEXT}} ``` Return side-by-side diff per paragraph, aggregate metrics, voice preservation check, and flagged-for-author items.

About this prompt

## Why most AI 'plain English' rewrites are flat They strip the voice along with the jargon. They turn warm, witty prose into AP-wire flatness. Or they go the other way and miss the actual hard work — leaving "utilize," "facilitate," and 38-word sentences in place because the model didn't know to look for them. The result is a rewrite the writer hates and the reader doesn't notice. ## What this prompt does differently It encodes the seven-pass plain-language discipline that government plain-language teams, top financial-disclosure editors, and Hemingway-grade copy editors actually use: length, words, verbs, voice, concretion, hedges, and reader-grade check. Each pass has specific moves and a measurable target. Voice and personality are firewalled — the prompt explicitly preserves wit, warmth, and idiom. ## The grade-level discipline Most AI rewrites improve clarity but don't measure it. The prompt estimates Flesch-Kincaid grade level before and after every paragraph, with an aggregate of 3+ grade-level drop as the success criterion. This makes the work auditable. ## The regulated-context safeguard In legal, medical, and scientific writing, some hedges and qualifiers are precision tools, not bloat. The prompt flags these for author decision rather than silently stripping them. "Generally," "largely," "in most cases" sometimes carry legal weight — the prompt knows when to pause and ask. ## What you get back - A side-by-side diff per paragraph (original / rewritten / key changes) - Per-paragraph and aggregate metrics: word count, sentence length, grade level, compression - A voice preservation check naming 3 sentences kept intact - Up to 5 places flagged for author decision - Up to 3 hedges left in because precision required ## Best for - Government and policy teams turning regulations into plain-language disclosures - Legal and medical writers producing patient-facing or consumer-facing documents - Financial communications teams writing proxy statements and 10-Ks for retail investors - B2B SaaS marketing teams cleaning up jargon-laden product copy - Academics translating research for the lay press without losing precision ## Pro tip Specify the target reading grade level explicitly. "Plain English" means grade 7-9 for general audiences but grade 12-14 for trade press. The right target is the reader's level, not the lowest possible level.

When to use this prompt

  • check_circleTranslating legal, medical, or financial disclosures into consumer-friendly language
  • check_circleCleaning up jargon-heavy B2B SaaS marketing copy without flattening voice
  • check_circleProducing audit-ready before/after diffs with measurable reading-grade reduction

Example output

smart_toySample response
Side-by-side diff per paragraph (original / rewritten / key changes with phrase-level edits), aggregate metrics (word count, sentence length, grade level, compression %), voice preservation check, and flagged-for-author items including precision hedges left intact.
signal_cellular_altintermediate

Latest Insights

Stay ahead with the latest in prompt engineering.

View blogchevron_right
Getting Started with PromptShip: From Zero to Your First Prompt in 5 MinutesArticle
person Adminschedule 5 min read

Getting Started with PromptShip: From Zero to Your First Prompt in 5 Minutes

A quick-start guide to PromptShip. Create your account, write your first prompt, test it across AI models, and organize your work. All in under 5 minutes.

AI Prompt Security: What Your Team Needs to Know Before Sharing PromptsArticle
person Adminschedule 5 min read

AI Prompt Security: What Your Team Needs to Know Before Sharing Prompts

Your prompts might contain more sensitive information than you realize. Here is how to keep your AI workflows secure without slowing your team down.

Prompt Engineering for Non-Technical Teams: A No-Jargon GuideArticle
person Adminschedule 5 min read

Prompt Engineering for Non-Technical Teams: A No-Jargon Guide

You do not need to know how to code to write great AI prompts. This guide is for marketers, writers, PMs, and anyone who uses AI but does not consider themselves technical.

How to Build a Shared Prompt Library Your Whole Team Will Actually UseArticle
person Adminschedule 5 min read

How to Build a Shared Prompt Library Your Whole Team Will Actually Use

Most team prompt libraries fail within a month. Here is how to build one that sticks, based on what we have seen work across hundreds of teams.

GPT vs Claude vs Gemini: Which AI Model Is Best for Your Prompts?Article
person Adminschedule 5 min read

GPT vs Claude vs Gemini: Which AI Model Is Best for Your Prompts?

We tested the same prompts across GPT-4o, Claude 4, and Gemini 2.5 Pro. The results surprised us. Here is what we found.

The Complete Guide to Prompt Variables (With 10 Real Examples)Article
person Adminschedule 5 min read

The Complete Guide to Prompt Variables (With 10 Real Examples)

Stop rewriting the same prompt over and over. Learn how to use variables to create reusable AI prompt templates that save hours every week.

pin_invoke

Token Counter

Real-time tokenizer for GPT & Claude.

monitoring

Cost Tracking

Analytics for model expenditure.

api

API Endpoints

Deploy prompts as managed endpoints.

rule

Auto-Eval

Quality scoring using similarity benchmarks.

Plain English / Hemingway-Grade Rewriter | AI Plain Language Editor Prompt | PromptShip