garrytan/gstack

Use Garry Tan's exact Claude Code setup: 23 opinionated tools that serve as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA

What it solves

gstack is designed to transform a single developer's experience with AI coding agents (like Claude Code) from a simple chat interface into a structured, virtual engineering team. It solves the problem of "AI slop" and the lack of of rigorous professional software engineering processes—such as strategic product rethinking, architectural review, QA testing, and security audits—when building software solo with AI.

How it works

It provides a comprehensive set of specialized "skills" (implemented as slash commands) that simulate different professional roles. These skills are organized into a a software development lifecycle (SDLC) pipeline: Think $\rightarrow$ Plan $\rightarrow$ Build $\rightarrow$ Review $\rightarrow$ Test $\rightarrow$ Ship $\rightarrow$ Reflect.

Each skill feeds into the next; for example, /office-hours creates a design doc that is then read by /plan-ceo-review and /plan-eng-review. The system integrates with various AI agents (Claude Code, OpenAI Codex, Cursor, etc.) and can drive a real web browser (via the Aside browser or a bundled Chromium instance) to perform live QA, scraping, and performance benchmarking.

Who it’s for

  • Founders and CEOs who are technical and want to ship products rapidly without a large team.
  • First-time AI coding agent users who need structured roles and frameworks instead of a blank prompt.
  • Tech leads and staff engineers who want to automate rigorous review, QA, and release processes on every pull request.

Highlights

  • Virtual Engineering Team: Includes 23 specialists (e.g., CEO, Eng Manager, Designer, CSO, QA Lead) accessible via slash commands.
  • Multi-Agent Support: Compatible with Claude Code and 9 other AI coding agents including Cursor and OpenAI Codex.
  • Multi-Step Pipeline: A structured process where each phase (Product $\rightarrow$ Design $\rightarrow$ Engineering $\rightarrow$ QA $\rightarrow$ Ship) feeds into the next.
  • Browser Integration: Ability to perform real-world browser testing, session management, and automated QA using a real browser environment.
  • Automated Documentation: Uses the Diataxis framework to automatically generate and update project documentation based on shipped features.

Related

  • Project
    rsmdt/the-startup

    A multi-agent AI framework for Claude Code that implements a spec-driven development workflow to turn ideas into shipped code via structured planning and parallel specialist agents.

  • Project
    deepset-ai/haystack

    An open-source AI orchestration framework for building production-ready RAG systems, autonomous agents, and LLM applications in Python.

  • Project
    soumatheusgomes/vibe-coding-toolkit

    A documented workflow and set of tools (plugins, prompts, lint rules, sub‑agent orchestration, memory handling) that let developers use Claude Code / Codex as a disciplined coding teammate. Includes a “Superpowers” discipline, parallel sub‑agents, quality‑gate ESLint config, and a full onboarding Playbook.

  • Project
    cwinvestments/memstack

    A structured skill framework for Claude Code that adds 131 professional skills for deployment, security, and development via an MCP skill loader.

  • Project
    echoVic/boss-skill

    An auditable agent-team workflow for coding agents that turns a single agent into a structured engineering team with event-sourced state and quality gates.