agentmods.dev

builders-stack CLAUDE.md

A set of instructions that maps a coding project’s folders, conventions, commands, and rules. The project is a monorepo, meaning one repository contains several related applications, services, and shared libraries.

In plain words
What is it for?
Use it when developing in builders-stack to find applications, services, libraries, and shipped packages, and to learn how to run or extend them.
Why use it?
It helps a coding agent understand where code belongs and which project rules it must follow before making changes. This reduces misplaced code and accidental breaks.

Instructions file

41 16d ago C scan 2,737 tokens original MIT
View the source file lonormaly/builders-stack
Install

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.

agentmods
npx agentmods add instructions/lonormaly/builders-stack/claude-md
Clone the repo
git clone --depth 1 https://github.com/lonormaly/builders-stack
Per session 2,737 This file is loaded in full into every session.
When invoked 2,737 The same file — it is already loaded in full.
Security scan C 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.02737 $0.02737
Opus 5 $0.01368 $0.01368
Sonnet 5 $0.00547 $0.00547
Haiku 4.5 $0.00274 $0.00274

Measured 2d ago against content hash 3e75bb0f3a7e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

builders-stack CLAUDE.md scanned grade C with 2 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

- **Third-party skills/MCPs — vet before you install:** a skill/MCP is code with your permissions + a payload the model obeys (the reason we swapped the SQL-injectable Postgres MCP for a read-only one). Before installing

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- **Third-party skills/MCPs — vet before you install:** a skill/MCP is code with your permissions + a payload the model obeys (the reason we swapped the SQL-injectable Postgres MCP for a read-only one). Before installing
CLAUDE.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — the map for coding agents

This is a bun-workspace monorepo. Read this before writing code; it tells you where everything lives so you don't reinvent what already exists.

Where things live

  • apps/ — public UI. apps/web (Next.js app), apps/landing (public marketing), apps/mobile (React Native / Expo — real starter).
  • services/ — anything with a URL. services/api (Hono + OpenAPI), services/ai-worker (background load), services/payment (Creem adapter).
  • libs/ — shared, never served. libs/ui (shadcn + tokens + Storybook), libs/auth (Better Auth), libs/db (Drizzle), libs/ai (Vercel AI SDK), libs/analytics (PostHog + Clarity + typed events), libs/email (Resend + React Email), libs/config (typed env), libs/api-types (shared API contract). Import by package name: @stack/ui, @stack/db, …
  • packages/ — what you SHIP (apps/services/libs are what you RUN). Distributable artifacts served to third parties: npm SDKs, embeddable widgets, CLIs. packages/widget (@stack/widget) is the worked example — an embeddable feedback widget with an IIFE (<script src>) + ESM build. Tag type:package; may depend on libs/* only; TERMINAL — nothing internal imports a package (shipped out, not consumed). No distributable to ship? Delete the folder.
  • ops/ — how you OPERATE it (the fifth bucket; apps/services/libs/packages are what the system is). Deploy · db · secrets · runbooks · local-CI. The outermost layer: it reaches down into the code to drive it (deploy, migrate, seed, provision) and nothing imports from ops/. Not a workspace, invisible to Nx — no build/lint/typecheck target, no boundary to violate. Secrets provisioning points to Ringtail and nothing else (ops/secrets/). Brand assets + strategy do not live here — those are docs/brand/. See ops/README.md.

Conventions (do not break)

  • No upward import: libs must not import from apps/services. Dependencies point down only.
  • One public door: import a lib from its package name / src/index.ts — never a deep path.
  • By feature, not by layer inside each app/service (billing/, not controllers/).
  • Every workspace extends the root tsconfig.base.json. Don't fork compiler options.
  • ORM is Drizzle (libs/db). One ORM only.
  • Payments go through the @stack/payment adapter interface — never call a vendor (Creem/Dodo/…) directly from an app. Swapping or adding a provider is a one-file change in services/payment/src/provider.ts; recipe in docs/stack/payments.md.

Read the full file on GitHub · 86 lines

Changes

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.

  1. 2d ago First seen · 86 lines · 2,737 tokens per session scan C 3e75bb0f3a7e

Subscribe to this mod's changes

builders-stack CLAUDE.md is an instructions file published in the GitHub repository lonormaly/builders-stack (41 stars, last pushed 16d ago), licensed MIT. It adds 2,737 tokens to every session, about $0.0137 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Securityai-agentsbetter-authbundrizzlehonomonorepo
Related

Other instructions, from other repositories

shipkit AGENTS.md

AGENTS.md instructions for chitamoor/shipkit, covering agents.md — ai coding guidelines, 1. project architecture, 2. auth conventions, 3. database conventions and 4. ai / mastra conventions.

chitamoor/shipkit · 3,842 tokens

create-better-t-stack AGENTS.md

AGENTS.md instructions for AmanVarshney01/create-better-t-stack, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and error handling conventions.

AmanVarshney01/create-better-t-stack · 938 tokens

nutrition-mcp CLAUDE.md

Instructions for akutishevsky/nutrition-mcp, covering claude.md, project overview, deploying, publishing to the registry and commands.

akutishevsky/nutrition-mcp · 11,933 tokens

Ornn CLAUDE.md

Instructions for ChronoAIProject/Ornn, covering claude.md — chrono-ornn, product positioning, tech stack, architecture and code standards.

ChronoAIProject/Ornn · 6,198 tokens

vobase CLAUDE.md

Instructions for vobase/vobase, covering vobase monorepo, packages, commands, quality rules and architecture.

vobase/vobase · 4,615 tokens

cldcde CLAUDE.md

Instructions for aegntic/cldcde, covering claude.md, commands, development, database and architecture.

aegntic/cldcde · 605 tokens