Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/vibestackdev/vibe-stack/claude-mdgit clone --depth 1 https://github.com/vibestackdev/vibe-stackWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01289 | $0.01289 |
| Opus 5 | $0.00645 | $0.00645 |
| Sonnet 5 | $0.00258 | $0.00258 |
| Haiku 4.5 | $0.00129 | $0.00129 |
Grade A, and why
vibe-stack CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — Vibe Stack Architecture Rules
For use with Claude Code (Anthropic's agentic CLI)
These rules prevent the most common AI hallucinations in Next.js 15 + Supabase + Stripe projects.
Project Overview
This is a Next.js 15 + Supabase + Stripe application using the App Router, TypeScript strict mode, and Tailwind CSS. All server-side auth MUST use getUser(), never getSession().
Tech Stack
- Next.js 15 (App Router only — no Pages Router)
- React 19
- TypeScript (strict mode)
- Supabase (auth, database, storage)
- Stripe (payments via Checkout Sessions)
- Tailwind CSS + shadcn/ui
- Zod (validation at every boundary)
Critical Build Commands
npm run dev # Start dev server
npm run build # Production build (catches type errors)
npm run lint # ESLint
npx tsc --noEmit # Type check without emitting
SECURITY RULES — NEVER VIOLATE
<security_critical>
-
NEVER use
supabase.auth.getSession()in server-side code (Server Components, Server Actions, Route Handlers, middleware). It reads the JWT from cookies WITHOUT verifying it — a forged token passes silently. ALWAYS usesupabase.auth.getUser()which makes a verification call to the Supabase auth server. -
NEVER import from
@supabase/auth-helpers-nextjs— it is DEPRECATED. Always use@supabase/ssrfor bothcreateBrowserClient(client) andcreateServerClient(server). -
NEVER put auth enforcement logic in
middleware.ts. Middleware runs on Edge Runtime and cannot securely verify Supabase JWTs. Middleware should ONLY callupdateSession()to refresh tokens. Auth enforcement MUST happen in layouts/pages viagetUser(). -
NEVER build a custom credit card form. ALWAYS redirect to Stripe Checkout. Custom forms create PCI compliance liability.
-
NEVER handle Stripe webhook events without calling
stripe.webhooks.constructEvent()first to verify thestripe-signatureheader. Without this, anyone can POST fake events to your webhook endpoint. -
NEVER upload files directly from client code using the Supabase anon key — it allows writing to any storage path. ALWAYS generate signed upload URLs server-side, scoped to
user.id.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 110 lines · 1,289 tokens per session scan A 6e8458e7f48c
vibe-stack CLAUDE.md is an instructions file published in the GitHub repository vibestackdev/vibe-stack (7 stars, last pushed 2d ago), licensed MIT. It adds 1,289 tokens to every session, about $0.0064 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
codeguide-starter-fullstack CLAUDE.md
Claude Code instructions for CodeGuide-dev/codeguide-starter-fullstack, covering claude code task management guide, documentation available, mandatory task management workflow, step 1: discover tasks (mandatory) and step 2: start each task (mandatory).
nextarter-tailwind AGENTS.md
AGENTS.md instructions for agustinusnathaniel/nextarter-tailwind, covering next.js: always read docs before coding, ai agent guidance: nextarter-tailwind, 1. mental model for agents, implementation delegation and tooling strategy.
expo_boilerplate CLAUDE.md
Instructions for chohra-med/expo_boilerplate, covering claude.md — mobilelauncher lt, 🌿 branches — the trunk is development, there is no main, project skill, what this project is and the master rules.
expo_boilerplate AGENTS.md
Instructions for chohra-med/expo_boilerplate, covering agents.md — mobilelauncher lt, project and 🎯 the master rules location.
mcp-quickstart AGENTS.md
Instructions for G12789/mcp-quickstart, covering agents.md, what this project is, layout, conventions and how to test changes.
guardvibe CLAUDE.md
Instructions for goklab/guardvibe, covering project guidelines, guardvibe security scanner, when user asks for security scan/fix, never do these and quick reference.