Agent skill pack · 30 skills
gstack
gstack is Garry Tan's set of slash-command skills that has a coding agent take on specialist roles across a sprint: product questioning, plan reviews, design, code review, real-browser QA, security audit and release. Each skill writes artifacts the next one reads, from design doc to test plan to PR. It suits founders and tech leads who want structured review and release steps around agent-written code.
Works with
- Claude Code
- Codex, Cursor, OpenCode, Gemini CLI and other agents
What problems it solves
- The agent builds the wrong thingAn office-hours session questions your framing, challenges premises and writes a design doc before any code exists.
- Bugs pass CI but break productionA pre-landing review reads the full diff, runs specialist and adversarial passes and auto-fixes only the obvious issues.
- Nobody actually clicks through the appThe QA skill opens a real browser or probes APIs and CLIs, reproduces each bug, writes a regression test and fixes it.
- Shipping steps get skippedThe ship workflow merges the base, runs tests and reviews, bumps the version and changelog, verifies and opens the PR.
The core workflow
These skills are built to be used one after another.
- 1
/office-hoursQuestions your idea in startup or builder mode, challenges premises, compares two or three approaches and writes a design doc, with no code. - 2
/plan-ceo-reviewReviews a plan as a founder would, in expansion, selective, hold or reduction mode, and asks before adding or cutting any scope. - 3
/plan-eng-reviewLocks a plan's architecture, data flow, edge cases and test coverage, with ASCII diagrams and a review report appended to the plan. - 4
/reviewReviews a branch diff for bugs that pass CI, dispatches specialist and adversarial passes, auto-fixes obvious issues and asks about the rest. - 5
/qaTests a site, API, CLI, job or webhook, reproduces each bug, writes a regression test first and commits each verified fix. - 6
/shipMerges the base branch, runs tests and reviews, bumps VERSION and CHANGELOG, verifies the frozen content, pushes and opens a PR. - 7
/land-and-deployMerges an approved PR, waits for CI and the deploy, then checks production health and reports the evidence. - 8
/canaryWatches a live URL after a deploy for console errors, failed pages and visual changes against baseline screenshots.
Skill guides
Explained step by step on this site
/csoRun a security audit that keeps evidence separate from claimsAudits code, infrastructure, supply chain or skills for exploitable defects, challenges each finding and can reproduce fixes in Docker.Read the guide →/office-hoursPressure-test an idea before your agent writes any codeQuestions your idea in startup or builder mode, challenges premises, compares two or three approaches and writes a design doc, with no code.Read the guide →/plan-ceo-reviewReview a plan for scope and ambition like a founderReviews a plan as a founder would, in expansion, selective, hold or reduction mode, and asks before adding or cutting any scope.Read the guide →/plan-eng-reviewLock architecture, edge cases and tests before you buildLocks a plan's architecture, data flow, edge cases and test coverage, with ASCII diagrams and a review report appended to the plan.Read the guide →/qaFind, reproduce and fix bugs with regression testsTests a site, API, CLI, job or webhook, reproduces each bug, writes a regression test first and commits each verified fix.Read the guide →/reviewCatch production bugs in a diff before it mergesReviews a branch diff for bugs that pass CI, dispatches specialist and adversarial passes, auto-fixes obvious issues and asks about the rest.Read the guide →/shipTake a branch from finished code to an open PRMerges the base branch, runs tests and reviews, bumps VERSION and CHANGELOG, verifies the frozen content, pushes and opens a PR.Read the guide →
All 30 skills
Skill names are shown as /name: depending on your agent, a skill runs when it is relevant or when you call it by name, and plugin installs may add a prefix. Highlighted skills have a guide on this site; the others open their folder on GitHub.
/autoplanRuns the CEO, design, developer experience and engineering reviews in sequence and brings only taste decisions to you.↗/benchmarkMeasures page load timings, Core Web Vitals and resource sizes in a real browser and compares them with a saved baseline.↗/browseDrives a real browser, preferring your signed-in Aside sessions, to open pages, click through flows, read the console and take screenshots.↗/canaryWatches a live URL after a deploy for console errors, failed pages and visual changes against baseline screenshots.↗/carefulWarns before destructive commands such as recursive deletes, dropped tables or force pushes, and hard-blocks the riskiest ones.↗/codexGets an independent review, challenge or consultation from the OpenAI Codex CLI and presents its output unedited.↗/csoAudits code, infrastructure, supply chain or skills for exploitable defects, challenges each finding and can reproduce fixes in Docker.→/design-consultationResearches your product space and proposes a full design system, written to a DESIGN.md file with font and colour previews.↗/design-htmlTurns an approved mockup into production HTML and CSS with computed text layout, detecting React, Svelte or Vue.↗/design-reviewAudits a live site for spacing, hierarchy and generated-looking patterns, then fixes them with atomic commits and before and after screenshots.↗/design-shotgunProduces several AI design mockups side by side in a browser board, collects your feedback and iterates until you pick a direction.↗/devex-reviewTests a developer product live by following its docs and getting-started flow, timing each step and comparing against the plan's scores.↗/diagramTurns an English description into a mermaid source, an editable Excalidraw file and rendered SVG and PNG, offline.↗/document-releaseChecks project docs against what changed in a release and updates the facts, asking about subjective edits.↗/freezeBlocks every file edit outside one directory you name for the rest of the session.↗/land-and-deployMerges an approved PR, waits for CI and the deploy, then checks production health and reports the evidence.↗/make-pdfConverts a Markdown file into a typeset PDF with optional cover, table of contents and watermark, or into HTML or Word.↗/office-hoursQuestions your idea in startup or builder mode, challenges premises, compares two or three approaches and writes a design doc, with no code.→/plan-ceo-reviewReviews a plan as a founder would, in expansion, selective, hold or reduction mode, and asks before adding or cutting any scope.→/plan-design-reviewScores a plan's design dimensions from zero to ten, explains what a ten looks like and adds the missing design decisions.↗/plan-devex-reviewReviews a developer-facing plan for onboarding friction, error messages and time to hello world, before any code is written.↗/plan-eng-reviewLocks a plan's architecture, data flow, edge cases and test coverage, with ASCII diagrams and a review report appended to the plan.→/qa-onlyExplores the same surfaces as QA and reports reproducible bugs with evidence, without changing code or tests.↗/qaTests a site, API, CLI, job or webhook, reproduces each bug, writes a regression test first and commits each verified fix.→/retroSummarises commits, work patterns and test health over a time window with per-contributor praise and growth areas.↗/reviewReviews a branch diff for bugs that pass CI, dispatches specialist and adversarial passes, auto-fixes obvious issues and asks about the rest.→/shipMerges the base branch, runs tests and reviews, bumps VERSION and CHANGELOG, verifies the frozen content, pushes and opens a PR.→/specInterrogates a vague request over several rounds and files a precise spec as a GitHub issue, optionally handing it to a new agent.↗/test-auditFinds tests that cost more than they protect, such as duplicates or assertion-free probes, and removes them only in approved batches.↗/verifyRuns the repo's own typecheck, lint, build and selected tests, fixes failures and repeats until green, without pushing.↗
No skill matches that filter.
Install
Pick your agent; commands come from the author’s README.
Claude CodeThe README wraps this command in a prompt you paste into Claude Code, which also adds a short gstack section to CLAUDE.md.
git clone --single-branch --depth 1 https://github.com/garrytan/gstack.git ~/.claude/skills/gstack && cd ~/.claude/skills/gstack && ./setupClaude Code (team mode for a shared repo)Run inside your repo; it commits a bootstrap so teammates get gstack with auto-updates, and needs Node.js for the required hook.
(cd ~/.claude/skills/gstack && ./setup --team) && ~/.claude/skills/gstack/bin/gstack-team-init required && git add .claude/ CLAUDE.md && git commit -m "require gstack for AI-assisted work"Codex, Cursor, OpenCode, Gemini CLI and other agentsUse --host auto for every detected agent or name one host; non-Claude hosts are listed as experimental.
git clone --single-branch --depth 1 https://github.com/garrytan/gstack.git ~/gstack
cd ~/gstack && ./setup --host auto # or: ./setup --host codex
./setup --status # one row per install: host, tier, scope, version, pathYou need Git and Bun 1.4.2 or later (plus Node.js on Windows); Claude Code gets the fullest support, while Codex, Cursor, Gemini CLI and others install through ./setup --host as experimental targets. On macOS the browser skills prefer the Aside browser with your signed-in sessions and otherwise fall back to a bundled Chromium that setup builds. Check the author’s README for changes before you run a command.
Questions
Is gstack free?
Yes. It is MIT-licensed, and the README states there is no premium tier or waitlist.
Does it work outside Claude Code?
Yes, through ./setup --host for Codex CLI, OpenCode, Cursor, Factory Droid, Kiro, GitHub Copilot CLI, Antigravity, Gemini CLI and others, but those are marked experimental and safety hooks there are advisory rather than blocking.
Does it send my data anywhere?
Usage telemetry stays off unless you opt in when asked. The update check contacts GitHub, and outside reviews send code to Codex or Claude; both can be turned off.
Which platforms are supported?
macOS and Linux get full support. Windows 11 works through Git Bash or WSL with minimal support, and needs both Bun and Node.js on PATH.
This page is trendhub’s own summary of a public repository. The skills, their text and every update live in garrytan/gstack by Garry Tan; we do not host or modify the files.