Prompt 1
Get set up.
Where: Paste into claude.ai (free web version)
You'll get: A personalized setup checklist, Claude Desktop or Claude Code installed and working.
This prompt makes Claude into your install partner. It figures out your OS,
recommends the right install path (Claude Desktop for non-technical;
Claude Code in terminal if you've used a command line), installs the
dependencies for whatever you want to build, and verifies everything works.
You are my AI install partner. I am non-technical and I want to start building real software with you. Walk me through getting set up on my machine, step by step.
1. Ask me which operating system I'm using (Mac, Windows, or Linux).
2. Based on my OS, recommend the simplest install path:
- On Mac or Windows: Claude Desktop (easiest, no terminal required)
- On Linux or for power-users: Claude Code via terminal
3. Give me the exact download/install commands or links. Verify the current canonical install command if you're unsure (Claude Code is at https://docs.claude.com/claude-code/setup).
4. Ask me what I want to build first (one sentence is fine). Based on that, tell me which dependencies I need to install (Ruby + Rails, Node + Next.js, Python, etc.) and walk me through installing them.
5. Help me get my Anthropic API key set up if I need one.
6. Create my first project folder and initialize git.
7. At the end, output a clean step-by-step checklist as markdown so I can save it for reference.
Ask me one question at a time. Wait for my answer before continuing. Be patient — I've never done this before. Use plain English, no jargon unless you explain it.
Prompt 2
Your first Claude Code prompt.
Where: Paste into Claude Code (terminal) or Claude Desktop after Prompt 1 is done.
You'll get: A scaffolded project with CLAUDE.md, git, and a "hello world" app running locally.
Once you're installed, this prompt initializes your real project. It picks
a stack appropriate to your idea, scaffolds the structure, sets up project
memory (CLAUDE.md), and verifies the app runs.
You are my AI co-builder. We're starting a new project together. Before you write any code, do this:
1. Ask me in plain English what I'm building, who it's for, and what the very first usable version needs to do (the smallest thing that could be considered "live").
2. Based on my answers, recommend a stack. Explain your tradeoffs in one paragraph.
3. Scaffold the project: create the folder structure, install dependencies, set up git, create a CLAUDE.md with the project context so future sessions remember what we're building, and add a sensible .gitignore.
4. Build a minimal working version that I can run locally. Just a "hello world" page or screen — enough to verify the stack is alive.
5. Run it. Tell me the URL or command to see it working.
6. End with a "what I just did" summary in plain English so I understand what's now in my project.
Move slowly. Ask me to confirm each major step before continuing. If something fails, debug it together — don't just rewrite.
Prompt 3
Deploy your first feature.
Where: Paste into Claude Code or Claude Desktop, in your project folder.
You'll get: A real, deployed feature anyone with the URL can visit.
This is the loop you'll use forever. Pick one feature, describe it in plain
English, build it, test it, deploy it. The moment your software stops being
a "maybe someday" and becomes real.
You are my AI building partner. We have a working project (from Prompt 2). Now we launch the first real feature.
1. Ask me to describe one feature in plain English — the smallest useful piece my real users would care about.
2. Write a one-paragraph spec back to me. Confirm I want to build that exact thing before continuing.
3. Build the feature. Test it locally.
4. Recommend a free or cheap host appropriate for my stack (Vercel for Next.js, Render or Fly.io for Rails, GitHub Pages for static, etc). Walk me through deploying it.
5. Give me the live URL.
6. Show me the exact loop to use next time: describe → build → test → deploy → improve. This is the pattern I'll repeat to add every future feature.
7. End with one specific next step I should do this week to keep momentum.
Make me run the commands myself. Don't hide what you're doing. I want to know what's happening on my machine so I can repeat it.