Skip to content

Latest commit

 

History

History
146 lines (99 loc) · 3.2 KB

quick-start.md

File metadata and controls

146 lines (99 loc) · 3.2 KB

Quick Start Guide

Get your project using Spec-driven development in 2 minutes

What You'll Learn

  • ✅ Install sce and adopt it in your project
  • ✅ Tell your AI about the methodology
  • ✅ Let AI work according to Spec-driven approach

Why This Is Better Than Ad-hoc AI Prompting

  • Specs reduce ambiguity before coding starts.
  • Orchestrate mode scales from single feature to multi-Spec parallel delivery.
  • KPI commands make progress and risk measurable, not just "looks done".

Step 1: Install sce (30 seconds)

npm install -g scene-capability-engine

Verify:

sce --version

Step 2: Adopt in Your Project (30 seconds)

cd your-project
sce adopt

This creates .sce/ directory with:

  • README.md - Project development guide
  • steering/ - Development rules
  • specs/ - Where Specs will live

Step 3: Tell Your AI (1 minute)

In your AI tool (Cursor, Claude, Windsurf, etc.), say:

Please read .sce/README.md to understand how this project works.

Your AI will learn:

  • This project follows Spec-driven development
  • Every feature starts with a Spec (requirements + design + tasks)
  • How to work with this methodology

Step 4: Start Working

Just ask your AI to implement features:

I need a user login feature with email and password.

Your AI will:

  1. Bootstrap a Spec draft (requirements/design/tasks)
  2. Refine requirements and design with you
  3. Run the staged Spec workflow
  4. Implement according to the Spec

If you want to run commands manually:

# Single-Spec recommended flow
sce spec bootstrap --name 01-00-user-login --non-interactive
sce spec pipeline run --spec 01-00-user-login
sce spec gate run --spec 01-00-user-login --json

# Multi-Spec flow (defaults to orchestrate mode)
sce spec bootstrap --specs "01-00-user-login,01-01-user-session" --max-parallel 3
sce spec pipeline run --specs "01-00-user-login,01-01-user-session" --max-parallel 3
sce spec gate run --specs "01-00-user-login,01-01-user-session" --max-parallel 3

The AI uses sce commands automatically - you don't need to learn them.

Optional Step 5: Verify Measurable Delivery (30 seconds)

Generate a sample KPI input file:

sce value metrics sample --out ./kpi-input.json --period 2026-W10 --json

Then run:

sce value metrics snapshot --input ./kpi-input.json --json

If you prefer manual input, use:

{
  "period": "2026-W10",
  "metrics": {
    "ttfv_minutes": 25,
    "batch_success_rate": 0.86,
    "cycle_reduction_rate": 0.34,
    "manual_takeover_rate": 0.16
  },
  "notes": "quick-start smoke run"
}

This gives you a machine-readable snapshot with risk level and output paths.

That's It!

You're now using Spec-driven development. Your AI understands the methodology and will follow it.

Key insight: You don't "use sce" - your project "follows Spec-driven methodology" and sce helps enforce it.

Next Steps

Version: 1.46.2
Last Updated: 2026-02-14