What is vocero-crm CLAUDE.md?

Instructions for coding agents working on Vocero CRM, a self-hosted WhatsApp customer-management application with an AI agent. It documents the technology stack, code layout, modification boundaries, and non-negotiable project rules.

Use it when developing, testing, or reviewing Vocero’s Next.js interface, PostgreSQL data layer, WhatsApp messaging, real-time events, AI features, Docker setup, or deployment configuration.

It gives agents the project’s architecture and operational constraints before they modify features, database code, authentication, messaging, or deployment files.

How kevinrivm/vocero-crm configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to kevinrivm/vocero-crm. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/kevinrivm/vocero-crm/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/kevinrivm/vocero-crm

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/kevinrivm/vocero-crm/claude-md/github.svg)](https://agentmods.dev/instructions/kevinrivm/vocero-crm/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/kevinrivm/vocero-crm/claude-md"><img src="https://agentmods.dev/badge/instructions/kevinrivm/vocero-crm/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/kevinrivm/vocero-crm/claude-md"><img src="https://agentmods.dev/badge/instructions/kevinrivm/vocero-crm/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 3,165 This file is loaded in full into every session.

When invoked 3,165 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.03165	$0.03165
Opus 5.5	$0.01266	$0.01266
Sonnet 5.5	$0.00633	$0.00633
Haiku 4.5	$0.00316	$0.00316

Measured 17d ago against content hash c1513aa0aeeb, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

vocero-crm 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 17d 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.

CLAUDE.md · 176 lines

How it starts

The opening of the file, as written. The whole thing — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Vocero CRM — Guía para Claude

Lee primero AGENTS.md: el mapa del código y las reglas de modificación viven ahí; este archivo solo añade lo específico de Claude Code.

Vocero es un CRM de WhatsApp open source (MIT), self-hosted, con el perfil y el conocimiento del agente y un Laboratorio de auto-evaluación. Desde la 2.0 no trae agente: contesta un cerebro conectado al lado (Nea u otro). Una instancia = un negocio. Este archivo guía a Claude Code (u otro asistente) para operar y modificar este repositorio — el caso típico: una agencia adaptando Vocero para un cliente.

Stack

Next.js 15 (App Router) + React 19 en monolito · TypeScript estricto (strict + noUncheckedIndexedAccess) · Tailwind CSS (sistema de diseño de la marca Vocero, el mismo de vocerocrm.com: tokens en src/app/globals.css, tema claro/oscuro, acento white-label por defecto #0d5bff, fuentes Archivo + Instrument Serif + IBM Plex Mono self-hosted vía next/font; el logo vive en src/lib/brand.ts y se dibuja con src/components/brand-mark.tsx) · PostgreSQL + Drizzle ORM (migraciones versionadas en drizzle/, aplicadas al ARRANCAR el contenedor) · Better Auth + plugin organization · Zod en todo input externo · nanoid con prefijos (ct_, cv_, msg_…) · pnpm · Vitest (unit) + guiones E2E en tests/e2e/ conducidos con Playwright · Docker multi-stage (standalone, healthcheck /api/health) · deploy en Coolify (Ruta A) o docker compose + Caddy (Ruta B).

Tiempo real por SSE (/api/events): heartbeat : ping ~25s, headers anti-buffering, catch-up por refetch con since=. Sin WebSockets, sin colas externas: el trabajo en segundo plano (el despacho al cerebro, el Laboratorio) es in-process.

Mapa del código (fronteras de modificación)

Quieres cambiar…	Toca…
El proveedor LLM (juez del Laboratorio y proxy del cerebro)	`src/lib/ai/` (adaptador OpenRouter-compatible, `chatJson<T>`; `provider.ts` resuelve Ajustes → IA o `OPENROUTER_*`)
El comportamiento del agente	No vive aquí: el CRM no trae agente (036). Lo que el agente lee —perfil, conocimiento (`renderKb` en `src/server/ai/prompts.ts`), `role`, `tools`, `llm`— sale por `src/server/bot/profile.ts`; lo demás es del cerebro (Nea)
Cómo piensa el cerebro con la llave del CRM	`src/server/bot/llm-proxy.ts` + `src/app/api/bot/llm/[...ruta]/route.ts`
Las personas o el juez del Laboratorio	`src/server/lab/personas.ts` · `src/server/lab/judge.ts`
El canal WhatsApp (Graph API)	`src/lib/meta/` (cliente único) + `src/server/whatsapp/`
Los canales opcionales (Instagram, Messenger; ADR-001)	`src/lib/channels.ts` (catálogo) · `src/server/channels/` (capacidades y bandera `CHANNELS`) · `src/server/instagram/` · `src/server/messenger/` · `src/server/zernio/` (transporte y firma de la API unificada, compartido)
Campos/tablas	`src/lib/db/schema.ts` → `pnpm db:generate` → migración nueva en `drizzle/`
La ingesta/envío de mensajes	`src/server/inbox/` (ingest idempotente, send con guard de sandbox, ventana 24h)
Cómo se identifica a un contacto	`src/server/inbox/identity.ts` (teléfono normalizado o `bsuid:<id>`)
Conectar el cerebro (Nea o el tuyo)	despacho: `src/server/dispatch/encolar.ts` (`despacharAlCerebro`) + `src/server/brains/` (`config.ts`: `despachoActivo`, `urlSaludCerebro`) · `src/app/api/bot/*` + `src/server/bot/auth.ts` (X-API-Key o Bearer) · contrato en `docs/openapi.json` · quién responde (avisos `sin_cerebro`, `apagado`): `src/server/bot/status.ts` + `GET /api/agent/brain-status`
La agenda (horarios, huecos, citas)	`src/server/agenda/` — detrás de la bandera `AGENDA` (`flag.ts`)
Cómo se entrega la reunión (Zoom, Meet…)	`src/server/agenda/connectors/` + catálogo en `src/lib/agenda-connectors.ts` · guía: docs/agenda-conectores.md
De qué anuncio llegó cada conversación (siempre visible)	`src/server/attribution/referral.ts` (normalización) · `creativo.ts` (copia de la imagen, solo hosts de Meta) · `store.ts` · tarjeta en `src/components/anuncio-origen.tsx`
Los números de Resultados (ventas, agente, origen y anuncios, higiene)	`src/server/analytics/` (un módulo por sección; periodo en la zona del negocio en `period.ts`; exclusión del Laboratorio en `shared.ts`) · contratos y tasas en `src/lib/analytics.ts` · UI en `src/components/results/` · spec 019
La atribución de anuncios y el reporte a Meta	`src/server/attribution/` — el `ctwa_clid`, la CAPI y Ajustes → Anuncios detrás de la bandera `ATRIBUCION` (`flag.ts`) + `src/lib/meta/capi.ts` · guía: docs/atribucion-capi.md
UI	`src/components/` + `src/app/(app)/`

Read the full file on GitHub · 176 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.

17d ago Changed · +7 lines · +280 tokens per session c1513aa0aeeb
29d ago Changed · +8 lines · +138 tokens per session ad3cd36e4580
1mo ago Changed · +2 lines · +201 tokens per session 1f33babc29ab
2mo ago Changed · +5 lines · +166 tokens per session da44989d79df
2mo ago First seen · 154 lines · 2,380 tokens per session scan A eb38cc0689af

Subscribe to this mod's changes

vocero-crm CLAUDE.md is an instructions file published in the GitHub repository kevinrivm/vocero-crm (419 stars, last pushed 4d ago), licensed MIT. It adds 3,165 tokens to every session, about $0.0127 per session on Opus 5.5. 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.

vocero-crm CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
vocero-crm CLAUDE.mdkevinrivm/vocero-crm	Instructions file	419	4d ago	MIT	A	3,165	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

vocero-crm: Instructions file for Claude Code