Borrowing it
Nothing to install: this file belongs to sumioshi/whatsapp-automation. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sumioshi/whatsapp-automation/main/CLAUDE.mdgit clone --depth 1 https://github.com/sumioshi/whatsapp-automationWrote 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.
[](https://agentmods.dev/instructions/sumioshi/whatsapp-automation/claude-md)<a href="https://agentmods.dev/instructions/sumioshi/whatsapp-automation/claude-md"><img src="https://agentmods.dev/badge/instructions/sumioshi/whatsapp-automation/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.
<a href="https://agentmods.dev/instructions/sumioshi/whatsapp-automation/claude-md"><img src="https://agentmods.dev/badge/instructions/sumioshi/whatsapp-automation/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.06912 | $0.06912 |
| Opus 5.5 | $0.02765 | $0.02765 |
| Sonnet 5.5 | $0.01382 | $0.01382 |
| Haiku 4.5 | $0.00691 | $0.00691 |
Grade A, and why
whatsapp-automation 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 — 328 lines — stays where its author put it; the contents beside it link to each section on GitHub.
WhatsApp deste projeto: <slug-do-grupo> (). Consulte o histórico via MCP
whatsapp-collector usando esse slug (ex: resumo_do_dia, ler_mensagens, buscar). Pra acompanhar
ativamente (ser acordado quando chega msg), arme um Monitor no data/<slug-do-grupo>/messages.jsonl
— ver "Vocabulário de acompanhamento" abaixo. Detalhes em .claude/whatsapp.json.
(Bloco gerado pelo /link-whatsapp; copie .claude/whatsapp.json.example pra começar.)
Vocabulário de acompanhamento (desambiguação)
"monitor" significava três coisas diferentes e isso já custou confusão. Mapa:
- "monitor" / "fica de olho" / "me avisa quando chegar mensagem" / "modo autônomo" → o operador
quer ser AVISADO/o agente quer ser ACORDADO quando chega mensagem nova. Isso é a tool Monitor
do harness, vigiando o arquivo de mensagens. Não é
alertar_chat. - "alerta" / "notifica no Mac" → notificação de tela pro humano (o operador) quando chega msg de
cliente. Tool MCP
alertar_chat. NÃO acorda o agente. - "a cada X min" / "puxa assunto sozinho" / "fica mandando mensagem" → re-executar um prompt em
intervalo. É o
/loop(ouCronCreate).
Recipe do Monitor de mensagem (sem zumbi)
Pra ser acordado quando chega msg nova num chat (modo monitor/autônomo), arme UM Monitor,
filtrando o que é seu (fromMe). O Monitor do harness expira em no máximo 30 min
(timeout_ms: 1800000; não existe mais persistent): na notificação de expiração, rearme com o
mesmo command. São duas receitas conforme a fonte:
Receita PADRÃO (nuvem, SSE — funciona com o Mac "limpo", sem coletor local): use a skill
global /plantao (.claude/skills/plantao/), que faz o fluxo completo — resolve slug, checa
e liga o watch do grupo (pegadinha: grupo novo nasce com captura desligada), mata zumbi e arma:
Monitor(timeout_ms: 1800000, command: scripts/plantao-stream.sh <slug> [<slug2> ...])
O script (scripts/plantao-stream.sh) encapsula o SSE de GET /api/events?slug=a,b + reconexão
COM retomada (Last-Event-ID, nada se perde no buraco) + lock por conjunto de chats (segunda
instância sai com código 3) + filtro: imprime Fulano: texto (recebida), Você (celular)/Você (web) (o operador respondeu na mão → contexto, não responda por cima) e IA da Meta (sua conta);
o que o próprio MCP/painel mandou (origem: coletor) não desce. --alertar acompanha todos os
chats com alerta ligado. Requer WAC_CLOUD_* no .env da raiz (ele mesmo carrega). Use path
absoluto se a sessão estiver fora do repo.
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 Changed · +236 lines · +5,364 tokens per session 84e7f788be96
- 2mo ago First seen · 92 lines · 1,548 tokens per session scan A e5f380dd58b9
whatsapp-automation CLAUDE.md is an instructions file published in the GitHub repository sumioshi/whatsapp-automation (2 stars, last pushed 7d ago), licensed Apache-2.0. It adds 6,912 tokens to every session, about $0.0276 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 2026-08-31.
whatsapp-automation CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| whatsapp-automation CLAUDE.mdsumioshi/whatsapp-automation | Instructions file | 2 | 7d ago | Apache-2.0 | A | 6,912 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 144k | today | MIT | A | 7,461 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,685 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 281k | yesterday | MIT | A | 4,335 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,086 | GitHub Copilot |