TechNomadCodeandClaude Opus 5.5
30b9b8cf55
- Rename prompt-templates/ to planning-prompts/ with numbered step folders, so GitHub lists them in workflow order. Folder, prompt file and skill now share one name per step (2-prd/prd.md is /prd). - Merge the full and ultra-lean MVP plan prompts into 5-mvp-plan/, as the /mvp-plan skill already did. - Rename the /mvp skill to /mvp-scope; the installer removes an earlier mvp install and tells the owner about the new name. - Rename agent-configs/claude-code-desktop/ to agent-configs/claude-code/, since the rules apply to the terminal too. - Rewrite each prompt guide in one short layout with previous and next links, drop outdated model and temperature advice, and fix the copied "final PRD draft" lines and the claimed v0.dev section. - Point to GitHub's Copy raw content button for copying a prompt. - Move the AI App Starter details from the root README into the starter guide, add the skills to the top table, and align the test plan wording with what the prompt produces. - Add issue forms, a Discussions link and a short contributing guide. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2.9 KiB
Planning prompts
Guided prompts that interview you, then write your plan. They work in any AI chat.
New here? Start with step 2, the PRD prompt. Do step 1 first if you have potential users you can ask.
| Step | Prompt | You bring | You get |
|---|---|---|---|
| 1. Research users | User research | A research brief, and later the responses | A questionnaire, then evidence for your PRD |
| 2. Define the product | PRD | Your notes, however rough | A product requirements document |
| 3. Map the experience | UX and user flow | Your PRD | Screens, user flows and interface specs |
| 4. Scope the first release | MVP scope | Your PRD, and UX specs if you have them | The first assumption to test, and what's in and out |
| 5. Plan the build | MVP build plan, full or lean | Your PRD and MVP scope | A development plan, or a short build spec |
| 6. Plan testing | Test plan | What you want to test | A test plan outline |
How they work
Most prompts tell the AI to write a document straight away, so it fills every gap with guesses. These make it interview you first:
- It asks before it writes. A few targeted questions at a time.
- It checks what it understood. It states its assumptions and confirms them with you before moving on.
- It drafts only once you agree. The document is built from your answers.
- Each step feeds the next. Later prompts take your earlier documents as input, so the plan stays consistent.
How to use a prompt
- Open the prompt file and copy it with the copy button at the top right of the file (Copy raw content).
- In a text editor, replace the placeholders, such as
[ <<< PASTE YOUR PRD HERE >>> ], with your material. Fill in or delete the bracketed lines underTONE & CONSTRAINTS. - Paste it into a new AI chat and answer its questions.
- When it offers to write the document, say yes. Review it and save it as a Markdown file: later steps ask for it.
Each folder's guide says exactly what to bring and what you get.
Tips
- Start a new chat for each prompt, and paste in the documents from earlier steps.
- Use a model that keeps a long conversation in view. Later steps paste in whole documents.
- Don't paste private information the AI doesn't need.
- The output is a draft. Review it before you rely on it.
Use them in Claude Code or Codex
Each prompt also installs as a skill, such as /prd. The skill runs the same prompt and fills it in from your message or files. See Skills.
When you're ready to build
These prompts work on their own, with any AI chat and any stack. To build the product with an AI coding agent, the AI App Starter sets up a Next.js, Supabase and Vercel project. It runs the PRD, UX and MVP scope prompts inside your agent, or uses the documents you already made with them.