SDStart Section 1

Living document

requirements.md

Version-controlled in this repository. All open items must be resolved (Clarification Gate) before implementation begins on a feature.

Project Vision

Build a personal learning platform that stores and organizes everything learned from the course Spec-Driven Development: From Vibe-Coding to AI Engineering. The platform must practice what it teaches: structured specs, executable intent, and zero ambiguity before building.

Clarification Gate

Zero unresolved items in requirements.md before implementation. Toggle items below to see the gate in action.

Gate closed — resolve open items first

2 open items remaining

Blocked
OI-001RESOLVED

Platform format: documentation-only repo, or evolve into a web app (React + Supabase)?

→ Evolved into this Next.js showcase app — docs + interactive learning UI.

OI-002OPEN

PRD vs sections: formal PRD per section, or section README sufficient?

OI-003OPEN

shape-spec workflow: specific CLI/tool, or manual conversation checklist?

User Stories

US-001Store course learnings by section

As a learner progressing through the SDD course,

I want each section's concepts, workflows, and examples stored in a structured, searchable format,

So that I can revisit foundations without relying on chat history or scattered notes.

Success criteria

  • Each completed section has a dedicated folder under docs/course/
  • Each section includes concepts, key terms, and practical takeaways
  • Section index stays current
US-002Maintain a living requirements document

As a spec-driven developer,

I want project requirements version-controlled alongside code,

So that intent is always traceable and the Clarification Gate can be enforced.

Success criteria

  • requirements.md exists at repository root
  • Open items tracked with status tags
  • Zero open items before feature implementation
US-003Use executable specifications

As a developer working with AI,

I want specifications detailed enough to generate tasks, plans, and prompts,

So that AI assistants execute intent instead of guessing.

Success criteria

  • User journeys documented for core flows
  • Acceptance criteria defined per feature
  • Data contracts specified where applicable
  • Templates available for reuse