Scaffold a Project

The scaffolding script creates a project structure with spec templates ready to go. No dependencies—just Python.

Quick start

curl -O https://waveform-analytics.github.io/spec-driven-site/scaffold.py
python scaffold.py

The script will ask for:

1. Project name — Used for the directory and in generated files
2. One-line description — What the project does
3. Project type — Affects which templates are included:
   • Data pipeline — Adds data flow architecture template
   • Analysis tool — General structure
   • Algorithm implementation — Adds algorithm and validation templates
   • General — Basic structure
4. AI tool — Creates the appropriate config file:
   • Claude Code → .claude/CLAUDE.md
   • Cursor → .cursorrules
   • GitHub Copilot → .github/copilot-instructions.md
   • Other/None → Skips AI config

What you get

• Directoryyour-project/

  • Directoryspecs/

    • spec-format.md How to write specs (reference this!)
    • 00-overview.md Your project overview (start here)
    • Directory01-requirements/

      • functional.md Requirements template
    • Directory02-architecture/

      • data-flow.md If you chose “Data pipeline”
    • Directory03-implementation/

      • …
    • Directory04-algorithms/ If you chose “Algorithm implementation”

      • algorithm-template.md
    • Directory05-validation/ If you chose “Algorithm implementation”

      • validation-plan.md
    • Directory99-decisions/

      • adr-000-template.md Decision record template
  • Directorysrc/

    • …
  • Directorytests/

    • …
  • .claude/CLAUDE.md If you chose Claude Code
  • CONTRIBUTING.md
  • README.md

The templates

Each generated file includes inline guidance explaining what to write. The guidance appears in HTML comments or bracketed placeholders:

<!--
This is your project's overview spec. It should answer:
- What problem does this project solve?
- Who is it for?
- What are the key capabilities?

Delete these instructions once you've written the overview.
-->

## Problem Statement

[What problem are you solving? Why does it matter?]

Fill in the brackets, delete the comments as you go.

After scaffolding

1. cd your-project

2. Open specs/00-overview.md and describe your project

3. Read specs/spec-format.md to understand how specs work here

4. Run git init to start version control

5. Start a conversation with your AI assistant about requirements

Next step

Once you have a project, write your first spec.