Core Principle

WHY → WHAT → HOW

Every layer's output feeds the next. Skip one and ambiguity explodes exponentially.

Phase 1
WHY

Requirements & Motivation

Understand the problem before solving it. Who needs this? What pain does it solve? What does "done" look like?

Key Outputs

  • User persona & pain point analysis (causal chains)
  • Scenario identification with priority (S01, S02...)
  • GIVEN/WHEN/THEN acceptance criteria per scenario
  • "Won't do" list to bound scope
Gate 1: Every P0 scenario has acceptance criteria
Output feeds input
Phase 2
WHAT

Product Design

Design the solution. For each scenario, define what users see, how they interact, and what state changes.

Key Outputs

  • Information architecture & navigation
  • Feature specifications per scenario
  • HTML prototypes (AI-generated)
  • UI-level GIVEN/WHEN/THEN (buttons, forms, states)
Gate 2: Every P0 scenario has interaction specs + prototype
Output feeds input
Phase 3
HOW

Implementation

Build the solution in 5 precise steps. Scenario-driven, test-first. AI generates code with full context.

0 Architecture overview & tech stack
1 Scenario → Sequence diagram → APIs emerge
2 API specs (OpenAPI YAML) → DB schema
3 Test case design (before code!)
4 AI code generation with full context
5 openlogos verify → automated acceptance
Gate 3.5: All tests pass, 100% coverage, AC traced

Why not just start coding?

When AI works without context, every decision is a guess. With 10 decisions per feature, that's 210 = 1,024 possible paths. Most of them are wrong. The three-layer model collapses this exponential space into one deliberate path.

Isn't this waterfall?

No. Waterfall means you finish all requirements before starting any design. OpenLogos works per scenario — you can take S01 through all three phases while S02 is still in Phase 1. The layers are a progression model, not a schedule.

Ready to try it?

Run openlogos init and let AI guide you through each phase.

Quick Start Next: Scenario-Driven →