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/aymericzip/intlayer/claude-mdgit clone --depth 1 https://github.com/aymericzip/intlayerWhat 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 | $0.00992 | $0.00992 |
| Opus 5 | $0.00496 | $0.00496 |
| Sonnet 5 | $0.00198 | $0.00198 |
| Haiku 4.5 | $0.00099 | $0.00099 |
Grade A, and why
intlayer 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 yesterday.
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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Intlayer: open-source i18n monorepo. Bun + Turborepo. Ships npm packages (React, Next.js, Vue, Svelte, Angular…) + hosted CMS (backend + dashboard).
Sub-CLAUDE.md per workspace:
apps/backend/CLAUDE.md— Fastify REST APIapps/app/CLAUDE.md— TanStack Start CMS dashboardpackages/@intlayer/design-system/CLAUDE.md— shared UI + TanStack Query hookspackages/@intlayer/api/CLAUDE.md— typed fetch clientpackages/@intlayer/core/CLAUDE.md— dictionary interpreter & transpilerpackages/intlayer/CLAUDE.md— main user package + CLI
Repo layout
apps/
app/ TanStack Start dashboard (CMS UI)
backend/ Fastify API server
website/ TanStack Start marketing/docs site (deployed)
showcase/ Showcase app
packages/
intlayer/ Main user package + CLI entry point
@intlayer/
core/ Dictionary interpreter, transpiler, formatters
config/ Config loader & types
api/ Typed fetch client for backend REST API
design-system/ Shared React UI + TanStack Query hooks
types/ Shared TS types (no runtime)
cli/ CLI implementation
editor/ Visual editor core (framework-agnostic)
editor-react/ React bindings for editor
lsp/ LSP implementation
mcp/ MCP server for IDE automation
babel/swc/webpack/vite/ Build-tool compiler plugins
react-intlayer/ React hooks & provider
next-intlayer/ Next.js integration
vue-intlayer/ Vue integration
(one package per framework)
utils/
ts-config/ Shared tsconfig base
tsdown-config/ Shared tsdown build config
plugins/
sync-json-plugin/ JSON sync plugin
sync-po-plugin/ PO file sync plugin
Commands (run from root)
Package manager: Bun.
bun run build # build all publishable packages + backend
bun run turbo build --filter=./apps/app
bun run turbo build --filter=./packages/react-intlayer
bun run lint # biome check
bun run lint:fix
bun run check # lint + format check
bun run check:fix
bun run typecheck # tsc --noEmit (sequential)
bun run test # vitest (concurrency=2)
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.
- yesterday First seen · 102 lines · 992 tokens per session scan A 8da2948e9ba0
intlayer CLAUDE.md is an instructions file published in the GitHub repository aymericzip/intlayer (816 stars, last pushed 2d ago), licensed Apache-2.0. It adds 992 tokens to every session, about $0.0050 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 2026-08-30.
Other instructions, from other repositories
lingo.dev CLAUDE.md
Instructions for lingodotdev/lingo.dev, covering extremely important instructions for claude, fyi, tools, testing and changesets.
web-worker-patterns AGENTS.md
Instructions for damiansire/web-worker-patterns, covering agents.md, estado actual (uno y uno), la regla de oro (no negociable) and gates antes de cerrar cualquier cambio.
web-worker-patterns CLAUDE.md
Instructions for damiansire/web-worker-patterns, covering claude.md, fases, convenciones de código and estándar nivel mundial.
platform AGENTS.md
Instructions for tutur3u/platform, covering agents.md - operating manual for autonomous ai assistants, 1. authority, 2. hard prohibitions, 3. mandatory actions and 4. repository map.
next-sanity-starter AGENTS.md
Instructions for backendforth/next-sanity-starter, covering agents.md — next-sanity-starter, tl;dr, principles (priority order — apply top-down when in doubt), repository map and the module pattern (the most important concept).
next-app-i18n-starter AGENTS.md
Instructions for s0vers/next-app-i18n-starter, covering agent guide, project, non-negotiables, cursor rules and deep reference.