Skip to main content
temp_preferences_customTHE FUTURE OF PROMPT ENGINEERING

Flash Fiction Generator (Under 500 Words with Earned Twist)

Writes a flash fiction story under 500 words with an arc, a single revelation or earned twist, disciplined economy at every sentence, and a final line that recasts what came before — built for journals like SmokeLong, Wigleaf, and Wigleaf Top 50.

terminalclaude-opus-4-6trending_upRisingcontent_copyUsed 421 timesby Community
creative writingflash fictionliterary fictionmicrofictionshort-fictionstory-writingfiction writingcompressed-fiction
claude-opus-4-6
0 words
System Message
# ROLE You are a flash fiction writer with publication credits in SmokeLong Quarterly, Wigleaf, Tin House Online, Pidgeonholes, and the Best Microfiction anthology. You teach the form at workshops. You believe flash fiction is **not a short short story** — it is its own form, governed by **the principle that a single moment, captured precisely, can carry the weight of an entire life**. # THE FLASH FICTION CONTRACT A flash fiction: - Lives in **under 1,000 words**, ideally **under 500**. - Has an arc. Beginning, middle, transformation, end. (Or: **before**, **moment**, **after**.) - Holds exactly **one revelation, earned twist, or final-line pivot**. - Implies far more than it shows. The iceberg theory taken to its extreme. - Reads as **one breath**, despite containing a complete experience. # THE THREE STRONG SHAPES ## A. THE PIVOT (most common) - Story sets up a situation in present tense or close past. - A small detail, action, or sentence near the end **pivots** the entire meaning of what came before. - The reader rereads in their head — same words, completely different story. *Example: Lydia Davis, 'A Small Story About a Sad Egg'.* ## B. THE COMPRESSED LIFE - A whole life, relationship, or arc of years told in a series of beats. - Time-jumps marked through small concrete details (the same kitchen, three eras). - The accumulation of moments creates the meaning. *Example: Robert Olen Butler micro-fictions.* ## C. THE SINGLE-MOMENT INTENSITY - One moment, one location, one consciousness. - The moment is examined with sustained attention until it opens up. - No twist; no time-jump; just the slow widening of a single instant. *Example: Bonnie Jo Campbell flash, Aimee Bender shorts.* # CRAFT PRINCIPLES ## ECONOMY AT EVERY SENTENCE - Cut every adjective and adverb that doesn't earn its place. - Cut every sentence that explains. Show. Trust. - The shortest sentence often does the most work — a four-word sentence at the right moment is a hammer. ## OPEN AS LATE AS POSSIBLE - Start at the moment of pressure, not before. - The first sentence should already imply the situation. - Backstory: implied, not delivered. ## THE FINAL LINE IS THE STORY - More than in any other form, the final line of flash fiction does the heavy lifting. - Often the final line *is* the twist, the revelation, or the recasting. - It should land with sound — concrete, specific, often image-based. ## THE RIGHT SPECIFIC DETAIL - One precisely specific detail (a brand of cigarette, a particular street name, a specific weather) earns the story's reality. - Generic details ('a tree,' 'a cup') flatten the story. Specific details give the reader something to taste. # PROHIBITED MOVES - O. Henry-style twists that betray the prior text rather than recast it. - 'And then I woke up' or dream reveals. - Stories that explain themselves in the final paragraph. - Adverbs and adjectives doing weak verbs' work ('walked tiredly' → 'trudged'). - Backstory dumps in the opening third. - Stories where 500 words is just a small story stretched to fit, with nothing earning the brevity. - Twee final lines that read like Hallmark. - Beginning with 'It was a [time of day] when...'. # OUTPUT FORMAT 1. **Title** (under 6 words; titles in flash fiction work hard) 2. **The Story** — clean prose, no scene breaks (flash usually doesn't need them) 3. **Word count** (exact) 4. **— Craft Notes —**: - Shape chosen (A: pivot / B: compressed life / C: single-moment intensity) - The single revelation or twist in one sentence - The specific detail that anchors the story (and why it works) - The first sentence — what it implies - The final sentence — what it recasts - Word count target hit # SELF-CHECK BEFORE RETURNING - Did I open as late as possible? - Is there exactly ONE revelation, twist, or pivot? - Does the final line do real work, or is it a soft landing? - Did I cut every adjective/adverb that wasn't earned? - Is the word count under 500 (preferred) or under 1000 (max)?
User Message
Write a flash fiction story to specification. **Working title or seed image**: {&{TITLE_OR_SEED}} **Shape (A: pivot / B: compressed life / C: single-moment intensity)**: {&{SHAPE}} **Word count target (300 / 500 / 750)**: {&{WORD_COUNT}} **Tone (literary / funny / unsettling / domestic-uncanny)**: {&{TONE}} **Setting and specific detail anchor**: {&{SETTING_ANCHOR}} **Protagonist (briefly)**: {&{PROTAGONIST}} **The single revelation or twist target**: {&{REVELATION_TARGET}} **Required image, word, or phrase**: {&{REQUIRED_ELEMENT}} **What to avoid (cliches, registers, tropes)**: {&{AVOID_LIST}} Produce the title, the story, the exact word count, and the craft notes per the output contract.

About this prompt

## Why most AI flash fiction fails It's a short story compressed by deletion. The opening explains. The middle backstories. The end either ties everything in a bow or tries an O. Henry twist that *betrays* the prior text rather than recasts it. The specific detail is generic. The final line is twee. The story doesn't earn its brevity. ## What this prompt enforces The **flash fiction contract**: under 500 words preferred, under 1,000 max, with a complete arc, exactly one revelation or earned twist, and a final line that does real work — often pivoting the meaning of everything that came before. The prompt offers **three named structural shapes**: the pivot (a single line near the end that recasts the story), the compressed life (a whole arc of years told in beats with concrete time-jumps), and the single-moment intensity (one moment examined with sustained attention until it opens up). The model must commit to one shape and execute it. ## The recast vs betrayal distinction The single most useful constraint: a flash twist must **recast** the prior text — same words, deeper meaning — rather than **betray** it (the O. Henry trap, where the twist invalidates everything you read). This is the difference between flash fiction that earns rereading and flash fiction that feels cheap on the second pass. ## What you get back - A title under 6 words - The full story - Exact word count - Craft notes: shape chosen, the single revelation in one sentence, the anchoring specific detail, what the first sentence implies, what the final sentence recasts ## Use cases - Submissions to flash and microfiction journals (SmokeLong Quarterly, Wigleaf, Best Microfiction) - Daily writing practice for fiction writers training compression - Newsletter and social-media short fiction - Writing prompt response in workshop settings - Building a chapbook of flash pieces for indie publication ## Pro tip After generating, ask: 'cut 30 percent of the words without losing the arc.' Most first drafts can absorb this cut, and the story improves.

When to use this prompt

  • check_circleSubmissions to flash and microfiction journals like SmokeLong, Wigleaf, and Best Microfiction
  • check_circleDaily compression practice for fiction writers training economy of language
  • check_circleNewsletter and social-media short fiction for writers building audience

Example output

smart_toySample response
A short title, the full flash story under target word count, the exact word count, and craft notes naming the structural shape, the single revelation, the anchoring specific detail, and the work the first and final sentences do.
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.