Free AI prompts.

I want to start building things with AI but I don't know what to make. Ask me 5 questions, one at a time, about what I do all week and what annoys me. Then suggest 3 things I could build that would actually be useful to me — ranked by how quickly I'd finish them.

Explain ___ to me twice. First in two sentences, as if I have never heard the word. Then again in a short paragraph, assuming I understood the first version. Don't use any other jargon to explain it.

I want to build ___. I'm a beginner using AI to write the code. Ask me 5 questions, one at a time, that will help you understand exactly what I want. Then write a one-paragraph description of the app we're going to build, simple enough that I could explain it to a friend.

Here's what I can already do with AI: ___. Here's what I want to be able to do: ___. Map the shortest path between the two as 5 steps. For each step, tell me what I'd build to prove I've got it.

I'm about to spend my weekend building ___. Before I start: what are the 3 most likely reasons I won't finish it? For each, tell me the smallest change to the plan that removes the risk.

Build the simplest possible working version of the app we just described. One HTML file, no installs, no frameworks. It must run when I open it in a browser. Explain in two sentences what you built before you show me the code.

Here's the error I got: ___. Don't just fix it — first tell me in one sentence what went wrong, as if you were explaining it to someone who has never coded. Then fix it and show me only the part that changed.

The app works but looks plain. Redesign the styling only — don't change what it does. Pick a colour palette that fits ___, make it work on a phone, and make the main button obvious. Show me the CSS and tell me the three choices you made and why.

I want to put this on the internet so I can send someone a link. Walk me through the simplest free way to do that, one step at a time. Wait for me to say 'done' before giving me the next step.

I want to add ___ to this project. Before writing any code: tell me which existing files you'd touch and what could break. Wait for me to say go. Then make the smallest change that works.

Go through this code and add a comment above each section explaining what it does in plain English — the kind a beginner would understand. Don't change any of the code itself.

Test this layout at 375px wide and tell me everything that breaks — text overflowing, buttons too small to tap, things running off the side. Fix them one at a time, showing me each fix before moving on.

This page feels slow. Find the 3 things most likely causing it, ranked by how much they cost versus how hard they are to fix. Don't fix anything yet — just tell me what you found and what you'd do.

Build a playable ___ game as a single HTML file — no installs, opens straight in a browser. Keep the rules simple enough to finish today. Add a score, a restart button, and make it work with both keyboard and touch.

Review this project the way a working developer would review a stranger's. Give me the 3 things most worth fixing, in order. Be blunt about anything that's actually broken, and don't compliment the parts that are just fine.

Here's how I'm planning to build ___: ___. Argue against this plan. What's the strongest case that I should do it a different way? Then tell me honestly whether that case beats my plan or not.

Here's a photo of my notes on ___. Make 10 quiz questions from it — a mix of easy and hard. Ask me one at a time, wait for my answer, tell me if I'm right, and keep score. At the end, tell me the two things I should revise.

I'm stuck on this: ___. Do NOT give me the answer. Ask me one question that would help me work out the next step myself. If I'm still stuck after three tries, give me a hint — still not the answer.

Here's my essay: ___. Don't rewrite any of it. Point out the 3 weakest paragraphs and tell me what's wrong with each — unclear argument, no evidence, waffle. Then ask me one question per paragraph that would help me fix it myself.

I have exams in ___ weeks covering these subjects: ___. I can study ___ hours on weekdays and ___ at weekends. Build me a week-by-week plan. Put the hardest topics earliest, build in one catch-up day a week, and tell me what to cut if I fall behind.

My teacher wrote this on my work: ___. Explain what they're actually asking me to change, in plain terms. Then give me one specific thing I could do differently next time.

I'm presenting ___ to ___. Here's my outline: ___. Play a sceptical audience member and ask me the 5 hardest questions I could get. One at a time — wait for my answer before the next one.

I want to make a short video about ___ for ___. Write a 45-second script. Open with the single most surprising thing about it — no 'hey guys', no intro. Mark where each shot changes.

Here's one thing I know well: ___. Turn it into 5 different posts — each making a different point, not the same point reworded. For each, tell me which platform it suits and why.

Here are three things I've written: ___. Describe my writing voice in 5 specific rules — sentence length, words I use, words I never use, how I open and close. Then rewrite this draft to follow them: ___.

My video is about ___. Give me 10 title options. Half should be plain and descriptive, half more curious. None should promise something the video doesn't deliver. Mark which two you'd actually pick and why.

Here are my messy notes on ___. Turn them into a 15-minute episode outline: an opening that isn't an introduction, 3 segments with a point each, and a closing line worth remembering. Flag anything I've claimed that I'd need to check.

Here's the post: ___. Write 3 caption options under 40 words. One states the point plainly, one asks something real, one tells a tiny story. No emoji walls, no 'drop a comment below'.

Here's how I spend a typical work week: ___. Identify the 3 tasks most worth handing to an AI tool, ranked by hours saved per week against how hard they'd be to set up. Be honest about which ones aren't worth it.

I want to test whether ___ is worth building. Design the smallest thing I could put in front of 10 real people this week to find out — not the full product. Then build it as a single working web page.

I'm about to ___. Before I do, give me the 3 strongest reasons this is a bad idea, ranked by how likely they are to actually bite me. Don't soften them, and don't finish with encouragement.

Here's a job ad: ___. Here's my CV: ___. Act as the hiring manager. Which 3 requirements am I weakest on, and what would make you put me in the 'no' pile in the first 20 seconds? Don't rewrite my CV — just tell me.

I'm interviewing for ___. Here's my background: ___. Ask me the 5 questions I'd least like to be asked, one at a time. After each answer, tell me in one line what was weak about it.

I need to tell ___ that ___. Write it in under 120 words. Say the difficult part in the first two sentences rather than burying it. No apologising twice, no 'just wanted to reach out'.

Here are my notes from a meeting: ___. Pull out only the decisions and the actions. For each action: who owns it and by when. If either is missing from my notes, list it separately as something I need to go and confirm.

I need to be usefully productive in ___ by tomorrow. Skip the history and the feature tour. Give me the 5 things I'll use 90% of the time, in the order I'll need them, each with one thing to try immediately.

My child is ___ years old and just built ___ using AI. Give me 5 questions I can ask them about it that would show whether they understand what they made, versus just copied it. Make the questions sound curious rather than like a test.

I'm considering an AI or coding class for my ___-year-old. Give me 8 questions to ask the provider that would separate a class where students build real things from one that's mostly slides and theory. For each, tell me what a good answer sounds like and what a bad one sounds like.

My child is ___ and has started using AI for schoolwork. Help me have one honest conversation about it. Give me 3 things worth saying, 2 things not worth saying, and the one question I should ask them and then be quiet.

My child used AI to ___ for a school assignment. Lay out the honest case that this is fine, and the honest case that it isn't. Then tell me the question I should be asking to decide — not the answer.

My ___-year-old spends their screen time on ___. Suggest 5 things they could build with AI that connect to that interest — not generic projects, actually connected. For each, say roughly how long it takes and what they'd have at the end.

My child just showed me something they made and I nodded without understanding it. Here's what it does: ___. Explain what's genuinely impressive about it and what's the easy part, so I know what to actually be proud of.

Look at the files in this folder. Suggest a folder structure that would make sense, and show me which files would go where — as a list, don't move anything yet. Flag anything that looks like a duplicate or that you're unsure about.

I'm choosing between ___ and ___. Don't tell me which to pick. Tell me the one question whose answer would settle it, and what each answer would imply. Then ask me that question.

___ days in ___, travelling with ___. We like ___ and hate ___. Build a day-by-day plan with no more than 2 fixed things per day. Mark which ones need booking ahead and which we can decide on the morning.

I'm selling these: ___. For each, write a short listing — what it is, honest condition, and one line on why someone would want it. Suggest a price range and say what I should photograph.

Here's a ___: ___. Tell me the 5 things in it that would most affect me, in plain English. Then separately list anything that looks unusual or worth asking about. Don't reassure me — I'll get advice, I just want to know what to ask.

Plan ___ dinners for ___ people. Constraints: ___. Reuse ingredients across meals so nothing goes off, keep at least two under 20 minutes, and give me one shopping list sorted by supermarket aisle.