All, Creative, Marketing, Business, Productivity, Coding, Education, Personal, Research, Other
All, GPT Image 2 (800), ChatGPT (274), Model-agnostic (247), Seedance 2.0 (242), Nano Banana Pro (226), Gemini Nano Banana (210), Gemini Nano Banana Pro (182), Grok Imagine (83), Claude (61), Gemini (44), Midjourney v7 (29), Nano Banana 2 (27), Veo 3 (25), Gemini Nano Banana 2 (24), GPT Image 1.5 (24), Midjourney (21), NotebookLM (20), Midjourney v8.1 (14), ChatGPT / Gemini (12), Kling AI 3.0 Turbo (12)
All, cinematic (198), photorealistic (115), editorial (106), portrait (96), commercial (39), fashion (38), minimalist (31), studio (31), luxury (30), product-photography (29), lifestyle (28), golden-hour (26), surreal (25), miniature (23), infographic (21), 3d-render (20), poster (20), candid (19), monochrome (19), macro (18)
All, Creative generation (2567), Template (299), One-shot (225), System prompt (50), Workflow (38), Persona (6), Agent (3)
65 prompts
Duel of the Fates lightsaber battle animation in HTMLCreativeThe prompt formula for photorealistic AI imagesA tested formula for prompting photorealistic AI images (subject, action, environment, lighting, camera, lens, perspective, composition, imperfections, mood) plus the winning example prompt from a five-round Tokyo street test.Research β
The Crucible: extract usable insight from contested or mystical sourcesA rigorous system prompt that reads a contested, foreign, mystical, ideological or overheated source and separates the real phenomenon and usable kernel from the metaphysical excess (the 'tailings'), with an assay, failure modes and per-session controls. For deep reading and study with Claude or ChatGPT.ProductivityPrompt Gauntlet: generate and rank 20 variations of any promptA meta-prompt that takes any prompt, extracts its intent, generates 20 standalone rephrasings, scores them and returns the top 5 with justifications. A self-contained prompt optimizer for ChatGPT or Claude.Business β
Pre-mortem prompt: stress-test a feature before you ship itA pre-mortem prompt that assumes a feature has already failed and works backwards to surface adoption, technical, second-order and business risks before you build it. Great for product decisions with ChatGPT or Claude.EducationThe 5 building blocks of a prompt that actually worksA framework for writing prompts that get good output on the first try, built from five parts most weak prompts are missing (goal, context, constraints, output format, examples), with a before/after example. Works with ChatGPT, Claude and any LLM.Personal β
Find the money leaking out of your budget (spending audit)A backward-looking personal-finance audit for ChatGPT or Claude: paste your income, spending and subscriptions and it hunts the leaks - forgotten subscriptions, duplicate charges, creeping prices - and totals the yearly savings. The Reddit OP found $2,400/year in about a minute.
Castle siege battle animation in a single HTML fileProductivity/STEELMAN code: make ChatGPT argue the opposite of your positionA command code that makes the model argue the strongest possible case for the opposite of your position whenever a message ends in /STEELMAN.Productivity/ATTACK code: make ChatGPT argue against what you just saidA command code you paste once at the start of a chat so that any message ending in /ATTACK gets argued against as hard as the model honestly can, instead of getting agreement.Productivity/NEXT code: make ChatGPT name the single most important next stepA command code that makes the model name the single most important next action and why it beats the alternatives whenever a message ends in /NEXT.Productivity/PLAINLY code: make ChatGPT drop the hedging and be bluntA command code that makes the model drop the hedging and give the blunt version of its answer whenever a message ends in /PLAINLY.Productivity/HOLES code: make ChatGPT surface your unspoken assumptionsA command code that makes ChatGPT or Claude point out the assumptions and gaps you have not stated out loud whenever a message ends in /HOLES.Productivity/ODDS code: make ChatGPT give an honest confidence levelA command code that makes the model state its honest confidence level, high, medium or low, and what would change it, whenever a message ends in /ODDS.Productivity/SOWHAT code: make ChatGPT cut the fluff and say why it mattersA command code that makes the model cut the fluff and explain why something actually matters whenever a message ends in /SOWHAT.Productivity/FAILHOW code: make ChatGPT predict how your plan goes wrong firstA command code that makes the model predict the most likely way a plan goes wrong before you commit to it, whenever a message ends in /FAILHOW.
3D flythrough of Hogwarts Academy in a single HTML file
Corporate lifestyle day-in-the-life video prompt
Romantic couple portrait on stone ledge prompt
Romantic couple hand-kiss travel portrait promptEducationDiagnose your weak spots on any topic (Claude)A Claude/ChatGPT prompt that runs an adaptive diagnostic quiz to surface your blind spots on a topic, then prescribes targeted exercises to fix them.
Build a Crossy Road clone in a single HTML file (three.js)
Neon heart reflection glasses portrait prompt
Compare Nano Banana 2 vs GPT Image 2 candid kitchen photo