C.R.A.F.T. is a structure for writing prompts detailed enough that an AI can't misread them — Context, Role, Action, Format, Target Audience. It's also, quite literally, how this site was built: the prompt below is what generated the master prompt behind every worked example in this library. Below: what each letter means, the C.R.A.F.T. prompt itself (copy it and use it on anything), and the actual result it produced for this site.
The situation the AI needs to understand before it can help.
The exact expertise the AI should act with — always a 20+ year expert.
A numbered, sequential list of steps — never one vague instruction.
Exactly how the output should be structured and presented.
Who the final output is actually for — shapes tone, depth, and language.
This is the meta-prompt itself — use it to turn any rough idea into a fully detailed, C.R.A.F.T.-structured prompt.
CONTEXT: We are going to create one of the best AI prompts ever written. The best prompts include comprehensive details to fully inform the Large Language Model of the prompt's: goals, required areas of expertise, domain knowledge, preferred format, target audience, references, examples, and the best approach to accomplish the objective. Based on this and the following information, you will be able write this exceptional prompt. ROLE: You are an LLM prompt generation expert. You are known for creating extremely detailed prompts that result in LLM outputs far exceeding typical LLM responses. The prompts you write leave nothing to question because they are both highly thoughtful and extensive. ACTION: 1) Before you begin writing this prompt, you will first look to receive the prompt topic or theme. If I don't provide the topic or theme for you, please request it. 2) Once you are clear about the topic or theme, please also review the Format and Target Audience sections below, and any example I provide. 3) If necessary, the prompt should include "fill in the blank" elements for the user to populate based on their needs. 4) Take a deep breath and take it one step at a time. 5) Once you've ingested all of the information, write the best prompt ever created. FORMAT: For organizational purposes, you will use an acronym called "C.R.A.F.T." where each letter of the acronym CRAFT represents a section of the prompt. Your format and section descriptions for this prompt development are as follows: -Context: This section describes the current context that outlines the situation for which the prompt is needed. It helps the LLM understand what knowledge and expertise it should reference when creating the prompt. -Role: This section defines the type of experience the LLM has, its skill set, and its level of expertise relative to the prompt requested. In all cases, the role described will need to be an industry-leading expert with more than two decades or relevant experience and thought leadership. -Action: This is the action that the prompt will ask the LLM to take. It should be a numbered list of sequential steps that will make the most sense for an LLM to follow in order to maximize success. -Format: This refers to the structural arrangement or presentation style of the LLM's generated content. It determines how information is organized, displayed, or encoded to meet specific user preferences or requirements. Format types include: An essay, a table, a coding language, plain text, markdown, a summary, a list, etc. -Target Audience: This will be the ultimate consumer of the output that your prompt creates. It can include demographic information, geographic information, language spoken, reading level, preferences, etc. TARGET AUDIENCE: The target audience for this prompt creation is a current large language model, such as Claude, ChatGPT or Gemini. If I provide an example, use it as a reference for your output. Again, take a deep breath and take it one step at a time.
This is the actual C.R.A.F.T.-structured prompt that came out of the framework above — and it's exactly what generated the site you're browsing right now: the taxonomy, the templates, and every worked example in the library.
CONTEXT uniquestoremarket.com is being built as a showcase-and-monetization site for the Self-Directed Learning Framework (SDLF) — an AI learning-agent prompt. Instead of promoting the framework directly, the site demonstrates its value through hundreds of concrete "worked examples": each one applies the SDLF prompt to a specific skill or topic and shows the resulting mini-course. Examples span multiple categories and difficulty levels, from simple, low-skill practical topics to university-level academic topics. Revenue comes from AdSense blocks placed between content sections throughout the site. Hosted on Cloudflare Pages via static file upload — no backend, database, or build step. ROLE A World-Class Expert with 20+ years of combined experience in content-hub information architecture, educational content design, and front-end development for high-traffic, ad-monetized static sites. ACTION 1. Map the visitor journey: home page → three featured examples → full library, organized by category and skill level. 2. Design the templates: home page, example detail page, library/category overview page — each with AdSense placeholder slots at natural breaks. 3. Propose a category and skill-level taxonomy that scales to hundreds of examples. 4. Write a first real batch of worked examples across categories and levels — full prompt-applied-to-topic, full resulting mini-course. 5. Insert clearly marked, empty ad-slot placeholders — no real AdSense code yet. 6. Build as static HTML with one shared CSS file and one vanilla JS file — no frameworks, no build step. 7. Make every page mobile-responsive and SEO-optimized: unique meta title/description, one H1 per page. 8. Review all content for originality and depth — avoid thin-content or AI-spam signals. FORMAT A complete, ready-to-deploy static file set: home template, example detail template, library template, shared CSS, shared JS, and the worked-out example pages — each file clearly labeled and upload-ready. TARGET AUDIENCE A broad range of visitors seeking free, self-guided learning — from people wanting simple, practical skills to people wanting structured, academic-level self-study — browsing many examples across categories and levels.
Every worked example in the library was built the same way — browse them all.
Browse the full librarySearch for "CRAFT prompt framework" and you'll find several acronyms that share the same five letters but not the same meaning. They're not wrong, just different. Here are the ones you're most likely to see, so you know which one someone means:
Which one should you use? The differences matter less than the habit they share: tell the AI the situation, who it should be, what to do step by step, and what the output should look like. This site uses Target Audience rather than Tone, because tone follows from knowing who the output is for. Tell an AI the reader is a nervous beginner and it adjusts tone, depth and vocabulary all at once. Tell it only "friendly tone" and it adjusts the tone alone.
One honest note. The meta-prompt on this page isn't an invention of this site. Versions of it circulate widely in prompt-writing tutorials. What this page adds is how it was applied: the prompt it produced, and the library that prompt built.
The same framework also produced GuideBook Agent, a structured prompt that builds a personalized learning guide with you instead of just explaining a topic — get the AI prompt for self-directed learning, or read the full Self-Directed Learning guide it's based on.