Garry Tan's gstack: An Open Source AI Agent Framework for Claude Code

✍️ OpenClawRadar📅 Published: March 17, 2026🔗 Source
Garry Tan's gstack: An Open Source AI Agent Framework for Claude Code
Ad

What gstack is

gstack is Garry Tan's open source software factory that transforms Claude Code into a managed virtual engineering team. It's MIT licensed and available on GitHub. According to Tan, he's written over 600,000 lines of production code in the last 60 days (35% tests) and does 10,000-20,000 usable lines of code per day as part of his CEO duties at Y Combinator.

How it works

The framework provides 13 specialist roles as slash commands, all using Markdown. These include:

  • /plan-ceo-review - Rethinks the product direction (e.g., expands "photo upload" to auto-identify products, pull specs and comps from the web, and draft listings automatically)
  • /plan-design-review - Catches AI slop patterns, provides design scores, exports DESIGN.md, flags patterns like gradient hero, icon grid, uniform radius
  • /plan-eng-review - Locks architecture with ASCII diagrams for data flows, state machines, error paths, plus test matrices and security concerns
  • /review - Finds production bugs, auto-fixes issues like orphan S3 cleanup, missing indexes, and flags race conditions

Other roles mentioned include QA lead (opens real browser and clicks through apps), release engineer (ships PR), and designer.

Ad

Example workflow

From the source: When requesting "photo upload for sellers," /plan-ceo-review responds with expansion proposals, /plan-design-review provides an 80-item audit and design system inference, /plan-eng-review creates ASCII diagrams and test matrices. After approval, Claude writes 2,400 lines across 11 files (models, services, controllers, views, migrations, tests) in about 8 minutes.

Technical details

The repository includes directories for workflows, bin, design-consultation, docs, document-release, gstack-upgrade, plan reviews, QA reviews, retro, review, scripts, setup-browser-cookies, ship, and test. Key files include ARCHITECTURE.md, BROWSER.md, CLAUDE.md, conductor.json, and package.json.

📖 Read the full source: HN AI Agents

Ad

👀 See Also

Claude DevTools: A Log Reader for Enhanced Claude Code Visibility
Tools

Claude DevTools: A Log Reader for Enhanced Claude Code Visibility

Claude DevTools is a local, open-source tool that reads Claude Code's existing log files in ~/.claude/ to provide detailed session visibility, including file operations with inline diffs, token breakdowns, context window visualization, and full subagent execution trees.

OpenClawRadar
Browser-native real-time coherence control system for Claude with SDE bands and Kalman filtering
Tools

Browser-native real-time coherence control system for Claude with SDE bands and Kalman filtering

A developer has built a real-time coherence control harness that runs entirely as a Claude artifact in the browser, treating conversation as a stochastic process with live Monte Carlo SDE paths, dual Kalman filtering, and behavioral signal detection.

OpenClawRadar
LLM Council Analysis Reveals Practical Claude Code Token Optimization Strategies
Tools

LLM Council Analysis Reveals Practical Claude Code Token Optimization Strategies

A developer used the LLM Council tool with 5 personas to analyze Claude Code usage patterns, identifying that extended thinking mode by default was the biggest token drain. The resulting playbook achieved 60-70% token reduction with same or better output quality.

OpenClawRadar
Layered Defense Framework for Claude Code Rule Enforcement
Tools

Layered Defense Framework for Claude Code Rule Enforcement

An IT operations professional built an 8-layer defense framework to enforce Claude Code rules after discovering that both CLAUDE.md prompts and blocking hooks could be bypassed. The approach adapts the Swiss cheese model from accident investigation to prevent workarounds.

OpenClawRadar