AI coding agents (Claude Code) often skip critical software engineering steps — they guess instead of planning, write code without tests, and present work without self-review. SDLC Wizard enforces a full software development lifecycle via hooks and skills: TDD, planning with confidence levels, self-review, CI shepherd, and cross-model review.
Software engineers using Claude Code who want engineering discipline enforced automatically — plan before coding, test before shipping, ask when uncertain. Built on 15+ years of SDET and QA engineering experience.
Original work by Stefan Ayala (@BaseInfinity). Inspired by battle-tested QA practices from production systems.
npx agentic-sdlc-wizard init
Then restart Claude Code. The wizard auto-detects your stack, generates CLAUDE.md/SDLC.md/TESTING.md/ARCHITECTURE.md, and enforces TDD + planning via hooks on every interaction.
Repository: https://github.com/BaseInfinity/agentic-ai-sdlc-wizard