What problem does it solve?
Production-grade software engineering framework that encodes Google-proven workflows, quality gates, and best practices for AI coding agents. Use when you need structured development lifecycle guidance: spec-driven development, incremental implementation, test-driven development, security hardening, performance optimization, or disciplined git/CI/CD workflows.
Core Features & Use Cases
- Spec-driven development and slash-command guided workflows ( /spec, /plan, /build, /test, /review, /code-simplify, /ship) to manage AI agent projects.
- End-to-end governance for production AI agents, including risk gating, verification, and CI/CD integration.
- Multi-LLM platform compatibility (Claude Code, Gemini CLI, Codex CLI, Cursor, Windsurf, OpenCode) to enable cross-platform workflows.
- Guardrails and best-practices inspired by Google engineering culture (trunk-based development, TDD, shift-left testing, and code-as-liability mindset).
- Clear collaboration roles and artifact ownership across spec authors, planners, builders, testers, and reviewers.
Quick Start
Start by defining a spec with /spec, then progressively plan, build, test, review, and ship.