v1 kits · rules · skills · tested starters
Stop prompting from scratch. Start shipping.
Plug-and-play AI coding kits (rules, slash-command skills and tested starter projects) generated into the native format of every major AI coding tool.
Browse kitsGet the free sampler
One-time purchase · v1.x updates included · free sampler needs no email
$ unzip kitquarry-saas-starter.zip && cd starter
# every tool's rules are already installed
├── .cursor/rules/ core.mdc security.mdc react-next.mdc …
├── CLAUDE.md (64 lines)
├── .claude/skills/ plan-feature review-changes security-audit …
├── AGENTS.md .agents/skills/
├── .windsurf/ rules workflows
└── .github/ copilot-instructions.md instructions prompts
$ npm ci && npm run check
✓ lint ✓ typecheck ✓ 15 tests ✓ build
> /plan-feature add team billing
wrote PLAN.md · 6 steps · acceptance checks definedWorks with
- Cursor
- Claude Code
- Codex
- Windsurf
- GitHub Copilot
The difference
Same prompt. Different output.
// app/api/webhook/route.ts
const body = await req.json()
// TODO verify signature
await db.orders.insert(body)
console.log("key", process.env.LS_SECRET)// verify HMAC over the raw body first
const raw = await req.text()
if (!verifySignature(raw, sig, secret))
return new Response("invalid", { status: 401 })
const event = WebhookSchema.parse(JSON.parse(raw))Illustrative example of what the rules ask for. AI tools can still ignore rules, so kits also make the agent verify its work.
How it works
Three steps, about two minutes
- 1
Pick a kit
Rules for your stack, or a full starter project with every tool's files already installed.
- 2
Drop in your tool's folder
Copy
cursor/,claude/,codex/,windsurf/orcopilot/to your project root. Rules load automatically. - 3
Ship with commands
Run
/plan-feature,/fix-bug,/review-changesand/security-audit. The agent verifies its work with real checks.
Kits
Built for how you ship
AI Coding Rules Pack
Battle-tested rules for security, testing, review and git across TypeScript, React/Next.js, Python/FastAPI and Node.
$19View kit →AI Agent Starter Kit
A tested TypeScript agent: tool-calling loop, provider adapter, MCP server template and evals, with agent rules for every tool.
$39View kit →SaaS Starter Kit
A minimal, tested Next.js 16 SaaS starter with auth, a database interface, a Lemon Squeezy webhook and a landing page. Lint, typecheck, tests and build all pass.
$49View kit →Godot Game Dev Kit
Godot 4 / GDScript 2.0 rules that stop AI writing Godot 3 and Python, with a tested starter project and headless checks.
$24View kit →Pricing
Pay once. Keep the updates.
AI Coding Rules Pack
$19 one-time
Battle-tested rules for security, testing, review and git across TypeScript, React/Next.js, Python/FastAPI…
DetailsGodot Game Dev Kit
$24 one-time
Godot 4 / GDScript 2.0 rules that stop AI writing Godot 3 and Python, with a tested starter project and…
DetailsComplete Vibe Coding Toolkit
$99 one-time
All four kits: Rules Pack, AI Agent Starter, SaaS Starter and Godot. $131 if bought separately.
Get the bundlePrefer to try first? The free sampler has 3 rules and 2 skills in every format.
Every tool, native format
- Cursor:
.cursor/rules/*.mdcwith description/globs/alwaysApply frontmatter. Workflows ship as manual @rules. - Claude Code:
CLAUDE.md(under 200 lines), path-scoped.claude/rulesand.claude/skills/*/SKILL.mdslash commands. - Codex:
AGENTS.md(well under the 32 KiB default cap) and.agents/skills. - Windsurf:
.windsurf/ruleswith trigger frontmatter (each file under 12,000 characters) and.windsurf/workflows. - GitHub Copilot:
.github/copilot-instructions.md,.github/instructions/*.instructions.mdwith applyTo, and.github/prompts/*.prompt.md.
Every file is generated from one master source and checked by a format validator, so the five versions never drift apart.
FAQ
Questions, answered
What is KitQuarry?
KitQuarry sells AI coding kits: rule files, skills/slash commands and tested starter projects for Cursor, Claude Code, Codex, Windsurf and GitHub Copilot. Each kit is generated from one master source into every tool's native format.
Which tools are supported?
Cursor, Claude Code, Codex, Windsurf and GitHub Copilot. Cursor gets .mdc rules, Claude Code gets CLAUDE.md + .claude/rules + skills, Codex gets AGENTS.md + .agents/skills, Windsurf gets .windsurf/rules + workflows, and Copilot gets copilot-instructions + instructions + prompt files.
What does 'tested' mean?
Every starter project passes its own checks before we zip it. The agent and SaaS starters pass lint, typecheck, tests and build from a clean npm ci. The Godot starter passes a headless parse, run and smoke test on Godot 4.3 and 4.7.2. Generated rule files are validated against each tool's documented format limits.
Is it a subscription?
No. Every kit is a one-time purchase, and v1.x updates are included.
What is in the bundle?
The Complete Toolkit ($99) contains all four paid kits: Rules Pack, AI Agent Starter, SaaS Starter and Godot Game Dev Kit. Bought separately they cost $131.
Is there a free version?
Yes. The free sampler has the core, security and agent-workflow rules, plus the /plan-feature and /review-changes skills, in every tool format.