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/felipenalves/invos/agents-mdgit clone --depth 1 https://github.com/felipenalves/InvOSWrote 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/felipenalves/invos/agents-md)<a href="https://agentmods.dev/instructions/felipenalves/invos/agents-md"><img src="https://agentmods.dev/badge/instructions/felipenalves/invos/agents-md.svg" alt="Measured on agentmods" 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 | $0.01340 | $0.01340 |
| Opus 5 | $0.00670 | $0.00670 |
| Sonnet 5 | $0.00268 | $0.00268 |
| Haiku 4.5 | $0.00134 | $0.00134 |
Grade A, and why
InvOS AGENTS.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 3d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — INVOS (sistema operacional do negócio)
Regras de operação multi-harness (Claude Code, Codex, Cursor, Grok, OpenCode, Gemini…).
CLAUDE.md só aponta pra cá (@AGENTS.md). Não invente regras fora deste arquivo + _memoria/.
Esse arquivo é editável. O /instalar complementa o final com regras do perfil — não apaga o boot abaixo.
Contexto do negócio
No início de toda conversa, ler (quando existirem e estiverem preenchidos):
_memoria/empresa.md— quem é o usuário, o que faz, como funciona o negócio_memoria/preferencias.md— tom de voz, estilo de escrita, o que evitar_memoria/estrategia.md— foco atual, prioridades, prazos
Usar como base pra qualquer resposta ou decisão. Ao sugerir prioridades, formatos ou abordagens, considerar o foco atual em estrategia.md.
Pra qualquer tarefa visual (carrossel, post, landing): consultar marca/design-guide.md como referência de estilo.
Não listar o que foi lido nem confirmar a leitura. Só usar o contexto.
Primeira vez
Se a memória ainda for placeholder / vazia → skill instalar antes de qualquer outra coisa.
Sessão cheia (opcional): skill abrir se existir e o user quiser boot ritual — o mínimo acima já basta.
Fluxo de trabalho
- Antes de executar: se existir skill relevante em
.agents/skills/(canônico), seguir a skill.
Claude Code também vê as mesmas skills via symlink em.claude/skills/. - Se não houver skill, executar a tarefa normalmente.
- Se a tarefa não tinha skill mas parece repetível (o usuário provavelmente vai pedir de novo):
"Isso pode virar uma skill pra próxima vez. Quer que eu crie?"
→ skill mapear-rotinas se sim.
Não perguntar em tarefa pontual ou pergunta simples. Só quando o padrão de repetição for claro.
Aprender com correções
Quando o usuário corrigir algo, melhorar uma resposta ou der instrução que parece permanente
(frases como “na verdade é assim”, “não faça mais isso”, “prefiro assim”, “sempre que…”, “evita…”, “da próxima vez…”):
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.
- 3d ago First seen · 135 lines · 1,340 tokens per session scan A da240d094d27
InvOS AGENTS.md is an instructions file published in the GitHub repository felipenalves/InvOS (5 stars, last pushed 11d ago), licensed MIT. It adds 1,340 tokens to every session, about $0.0067 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
honcho CLAUDE.md
Claude Code instructions for plastic-labs/honcho, covering claude.md, honcho overview, what is honcho?, core concepts and peer paradigm.
Dragon-Brain CLAUDE.md
Instructions for iikarus/Dragon-Brain, covering dragon brain — claude.md, the harness, audit remediation (april–may 2026, complete 2026-05-09), the lie this audit closed and the contract that matters now.
brainlayer AGENTS.md
Instructions for EtanHey/brainlayer, covering brainlayer — a letter from etan, brainlayer agent notes (operational), review guidelines, key paths and database.
mnemonic memory-files.instructions.md
Instructions for modeled-information-format/mnemonic, covering memory file guidelines (mif level 3), required frontmatter fields, optional temporal fields, optional provenance fields and memory types.
dakera-cli CLAUDE.md
Instructions for Dakera-AI/dakera-cli, covering dakera-cli, key commands, architecture and conventions.
cosmos-mcp CLAUDE.md
Instructions for teampolarity/cosmos-mcp, covering cosmos-mcp, layout, tools (authoritative list in src/tools/index.ts), cli subcommands and imessage is the headline source.