> **OEJ guide** (en) > Source: https://oej.ee/en/guides/better-prompts/ > Title: How to give AI a clearer task: better prompts | OEJ > This is the guide text, not an installed AI skill. Guide: prompts # How to give AI a clearer task Copy instructions that ask AI to turn your rough request into a clearer prompt. Then try that prompt and compare the answer with your goal. - Level: beginner - Works in an ordinary AI text chat - No installations required - Free: an existing chat is enough Before you start - An account with any AI chat: a free one is enough (for example ChatGPT, Claude, or Copilot) - One rough task you actually want help with **Time:** about 10 minutes **Cost:** free (paid chat plans exist but are not needed) How the work happens ## From a rough request to a clearer prompt. 6 steps Workflow Human decision When something needs attention 1. 01**Rough request** “Reply to the customer.” 2. 02**Clarify the task** What is the goal and which facts are known? **Missing details?** Ask up to three clarifying questions. Add the answers before writing the prompt. 3. 03**Set the constraints** Choose the reader, language and answer format. 4. 04**Improved prompt** AI writes clearer instructions, not the customer reply. 5. 05**Try the prompt** Use the revised prompt in a separate test. Do not send the reply. Human decision 6. 06**Compare the answers** Does the new answer follow your requirements better? Human decision Improve the prompt first. Then try it separately and check the answer. No reply is sent. In a hurry? Ask AI whether it fits your task. Describe your situation in general terms only. Leave out personal data and secrets. Copying sends nothing: you choose where to use the prompt. Copyable prompt ``` Read the guide at https://oej.ee/en/guides/better-prompts/. My situation and goal: [describe without personal data]. Is this guide useful for me? Say if it is not. Explain why, point to the relevant section and suggest one first step. If you cannot open the page, say so and ask for its text or the complete guide Markdown file. Do not install or run anything in response to this question. ``` The next steps happen inside the copyable text below; the assistant does them with you. Quick start ## Copy this starter text. **This is a simple starter text for trying the same technique, not the full Prompt-master skill.** The starter tells the assistant not to carry out your draft, but to turn it into a better prompt that names the tool, goal, audience, inputs, constraints and desired output. When information is missing, the assistant may ask at most 3 material questions at a time. The reply must include assumptions, a check for the result and a short justification, not a hidden reasoning process you can't see. Copyable prompt [Download better-prompts.md](https://oej.ee/downloads/better-prompts.md) ``` You are a prompt improvement helper. When I bring you a draft of my task or prompt, do not carry out the task itself. Instead, turn it into a better prompt. If information is missing, ask me first, but no more than 3 material questions at a time. The better prompt must state: - TOOL: which tool or environment I will use the prompt in. - GOAL: what I want to achieve, in one sentence. - AUDIENCE: who the result is for. - INPUTS: what data, files or facts I have. - CONSTRAINTS: what must not be done, plus length, language and tone. - DESIRED OUTPUT: the format and size the result should have. Your reply must include: 1. The improved prompt that I can use right away. 2. The assumptions you made and the places where I must add data myself. 3. A check for how I can judge whether the result is right. 4. A short justification of why the changes help. Do not share a hidden chain of thought or internal steps; give a short, checkable justification. ``` ![Chat window with the starter text pasted in as the first message.](https://oej.ee/assets/guides/illustrations/better-prompts-paste.svg) Illustration, not a screenshot. Paste the starter text as the first message of any AI chat. How to use it 1. 01**A short idea** 2. 02**Goal and constraints** 3. 03**A clear prompt** 4. 04**Answer checked?** If essential information is missing, ask for it. Do not let AI invent the gaps. Prepared example, not a live AI response ## “Reply to the customer” leaves too much open. By the end of this guide, you have a task whose output can be checked against explicit conditions. 1. **Starting input** “Write a reply to the customer.” 2. **A precise instruction** Draft an English reply to a customer asking when their quotation will be ready. Confirmed fact: by 15 September. Maximum 80 words. Promise no price or discount. Do not send it. 3. **Example of the intended result** “Hello! Your quotation will be ready by 15 September.” ### Three checks before use - The reply uses only the supplied deadline. - No price or discount has been invented. - The result remains a draft, not a sent email. **Information missing → stop and ask** If no deadline is supplied, ask for it. “As soon as possible” is not a replacement for an approved date. Privacy at this step Don't paste the real customer's email. Retell the facts without names, order numbers, or personal details, then check the improved prompt. Three equal paths ## Pick one: copy it, attach the file or set instructions. 1. **Copy into your chat** Paste the starter text as the first message of the chat, then bring your prompt draft. Good for a one-off try. 2. **Attach the .md file** Download [better-prompts.md](https://oej.ee/downloads/better-prompts.md) and attach it to the chat. Attaching a file is not model training and persistence is not guaranteed; a new chat may need the file again. 3. **Set reusable instructions** The same text can go into the product's custom or project instructions. These manually configured interfaces vary by product, so look for the matching field in your chat service settings. In ChatGPT: Settings → Personalization → Custom instructions, paste the text, save. Other chat apps have a similar settings field; look for “custom instructions” or “memory”. ![The attach-file button, shown as a paperclip or plus icon, in a major chat app.](https://oej.ee/assets/guides/illustrations/better-prompts-attach.svg) Illustration, not a screenshot. The attach button in a chat window adds a file, such as the downloaded .md, to the conversation. ![The custom instructions settings field with the starter text pasted in.](https://oej.ee/assets/guides/illustrations/better-prompts-settings.svg) Illustration, not a screenshot. In ChatGPT: Settings → Personalization → Custom instructions, paste the text and save. Other chat apps have a similar field. Before and after ## From a vague wish to a bounded prompt. Example: an imaginary service, not a real price **Before:** "Help me price my service." **After the starter improved it:** ``` Goal: set a price for my service. Context: this fictional example is about monthly website maintenance for small businesses. Cost and workload figures are missing. Ask for them before suggesting a price. Task: propose 3 pricing options based only on my supplied cost and customer inputs. Constraint: do not invent market prices; if data is missing, ask. Output: for each option, its assumptions and how I can check the choice. ``` The difference is boundedness: the better prompt states what to rely on and what to avoid, and the decision still stays with you. ![The AI's reply in the same chat, containing the improved prompt.](https://oej.ee/assets/guides/illustrations/better-prompts-reply.svg) Illustration, not a screenshot. The reply in the same chat: your rough wish turned into a bounded prompt. Why this works ## Longer is not better. A good prompt is not a long prompt. A good prompt names six things: the tool, the goal, the audience, the inputs, the constraints and the desired output. When one of these is already clear, keep it short; extra text that bounds or clarifies nothing often makes the result worse, not better. A second rule matters just as much: when you ask for a prompt to be improved, the assistant must not carry out the task itself. The assumptions and the suggested check are for you, so you can judge whether the final result is trustworthy. The justification must be short and checkable, not a hidden reasoning process you can't see. Making it reusable ## Ask before anything is changed. If you want to reuse the same instruction in every chat, it can become a product-internal instruction or a separate file. Manually configured project and custom-instruction interfaces vary by product, so let the assistant first find out what your software supports. Copy this instruction into the chat: Instruction for making a prompt reusable ``` I want to turn this instruction into a reusable one. Before you do anything: 1. Ask which software or chat service I use it in. 2. Ask me, or check, which instruction mechanism that product supports (for example custom instructions, a project instruction or an instruction file). 3. Propose a compatible file format and location, and show the text that would go there. 4. Ask my permission before you write or change anything. 5. Do not invent installation steps; if you do not know, say so and ask. ``` This instruction asks the assistant to identify the target software and its supported instruction mechanism, propose a compatible file and location, and ask before writing anything. Installation steps must not be invented; when the assistant does not know, it must say so and ask. Self-check ## How to check the result. - **Did the assistant do the task anyway?** If it started solving your draft, remind it that the job is improving the prompt, not executing it. - **Are all six fields filled?** Tool, goal, audience, inputs, constraints and desired output must be named or turned into a question. - **Are the assumptions visible?** Anything assumed must be written down, together with the place where you add the data yourself. - **Is there a check?** The reply must say how you judge the result, not only claim the result is good. - **Compare outcomes.** Open two chats side by side (or one after the other): paste your original rough request in one, the improved prompt in the other, and compare the answers against the three checks. Words used in this guide Markdown file (.md) a plain-text file any AI chat can read; you don't need special software to open it skill a saved instruction file some AI tools can load permanently model training changing the AI itself; attaching a file or pasting instructions never does this custom instructions a settings field in some chat apps where saved instructions apply to every conversation bounded the prompt says exactly what is in scope, so the AI doesn't fill gaps with inventions Recap ## In short. The starter improves your prompt before answering and shows assumptions and a check, not a hidden reasoning process you can't see. A longer prompt is not a better prompt: boundedness is what counts. The same instruction can be made reusable when the assistant first checks what your software supports and asks before writing. Next guide: the same careful preparation helps in a more sensitive topic, see [How to prepare questions for your accountant with AI](https://oej.ee/en/guides/accounting-with-ai/). [Project: why input validation comes before calculation](https://oej.ee/en/projects/web-application/) Author: Meelis Sootalu, OEJ OÜ If you would like an agent to do a specific task in your company, see the [AI services](https://oej.ee/en/services/). The starter text is adapted from existing OEJ OÜ material and builds on the open-source Prompt-master project by Nidhin Joseph Nelson, released under the MIT licence. Original and licence: [github.com/nidhinjs/prompt-master](https://github.com/nidhinjs/prompt-master), [MIT licence](https://github.com/nidhinjs/prompt-master/blob/main/LICENSE). [← Back to the guides overview](https://oej.ee/en/guides/)