/review finds bugs that pass CI but blow up in production. Staff engineer-level analysis with auto-fix. /codex adds independent cross-model second opinion for critical code.
/qa tests your app with live browser, finds bugs, fixes with atomic commits. /cso security audit covers OWASP Top 10 + STRIDE threat model with zero-noise design and exploit scenarios.
/design-review runs the same audit as plan-design-review, then applies fixes. Before/after screenshots, atomic commits, visual documentation. /devex-review audits live developer experience.
gstack is Garry Tan's open-source AI coding assistant framework with 23 specialist skills and 8 power tools. It turns Claude Code into a virtual engineering team—CEO, Designer, Eng Manager, Release Engineer, QA Lead, and more. This series covers the full sprint workflow: Think → Plan → Build → Review → Test → Ship → Reflect.
Step-by-step installation and setup of Superpowers on Claude Code, Cursor, Codex, and OpenCode—covering how it works, how to verify it’s active, and how to troubleshoot the most common "not working" issues.
writing-skills teaches you how to turn a workflow into a reusable skill: structure, triggers, examples/counterexamples, and "documentation TDD" to validate that the skill is actually executable. Includes templates and review checklists for building your own Superpowers skill library.
brainstorming is a core Superpowers skill for turning vague ideas into clear requirements and a design spec. It forces an AI coding assistant to clarify goals, constraints, and tradeoffs before writing code. This article covers triggers, outputs, and common anti-patterns.
writing-plans turns a design doc into an implementable plan: atomic tasks (2–5 minutes each), explicit acceptance criteria, and verification steps. This prevents AI coding assistants from jumping into huge unreviewable changes and helps you ship in small, verifiable increments.
test-driven-development is a core Superpowers skill that enforces the RED→GREEN→REFACTOR loop. This guide shows how to do TDD with an AI coding assistant: test-first constraints, good test granularity, common anti-patterns, and practical workflows you can reuse.
systematic-debugging is a Superpowers skill for root-cause analysis. It enforces a four-phase workflow (reproduce → hypothesize → validate → fix) so you stop guessing and start proving. This guide shows what evidence to collect and how to prevent regressions.