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/fhdumont/code-ai-template/claude-mdgit clone --depth 1 https://github.com/FHDumont/code-ai-templateWhat 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 | $0.01127 | $0.01127 |
| Opus 5 | $0.00563 | $0.00563 |
| Sonnet 5 | $0.00225 | $0.00225 |
| Haiku 4.5 | $0.00113 | $0.00113 |
Grade A, and why
code-ai-template 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 — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
AGENTS.md é a instrução canônica — leia primeiro, tudo lá vale aqui. Este arquivo só mapeia a mecânica do loop pros botões do Claude Code.
Os rituais são skills
Os três momentos do loop têm skill própria em .claude/skills/ — é a forma preferida de acioná-los, porque a ordem dos passos e as pausas já vêm dentro:
| Skill | Quando | O que faz |
|---|---|---|
/planejar-fase |
abrindo o plan mode | inbox de achados → docs vivos quentes → auditoria do código → grilling → spec em docs/SETUP.md |
/fechar-fase |
fim da fase em code mode | critérios → revisão de contexto fresco → passada de docs → pausa pro dono → PR/merge/limpeza |
/verificar-referencia |
de tempos em tempos | novidades dos repos de docs/reference/referencias.md → issues achado → SHA avançado |
A quarta, grilling, é a primitiva model-invoked por trás do /planejar-fase — interroga o dono em rodadas até a fronteira de decisões esvaziar. Chame direto quando quiser estressar um raciocínio fora do ritual.
Mecânica no Claude Code
- Plan mode = a passada de raciocínio. Nele você audita o código, discute o aberto e redige a spec no
docs/SETUP.md.ExitPlanMode(aprovação do plano) = aprovação da spec. Não planeje o trivial. /clearentre os modos (não/compact): aprovada a spec, o dono limpa o contexto e a execução roda numa sessão nova, lendo só oSETUP.md.- Inbox de achados no início do plan mode:
gh issue list --label achado --state openantes de auditar o código. Triagem e fecho de issue seguemAGENTS.md§Inbox de achados. - Subagentes =
Task. A revisão de fecho de contexto fresco (AGENTS.md§O loop) roda como subagente viaTask, com só os docs vivos + o diff; etapa marcadadelegar a subagentena spec também. - Fecho da fase pela CLI
gh. A sequência inteira — critérios, pausa pro dono, PR, merge, limpeza — está emAGENTS.md§Git, e o/fechar-fasea executa na ordem. O dono não toca no GitHub. - Git destrutivo é bloqueado por hook.
.claude/hooks/block-dangerous-git.sh, registrado como PreToolUse/Bash em.claude/settings.json, barrapush --force,reset --hard,clean -f,branch -Decommitcom o branch atual emmain— as regras deAGENTS.md§Git viradas guardrail duro. A mensagem do bloqueio traz a rota certa; siga por ela ou peça ao dono, sem contornar o hook. - Prompts curtos: o dono aciona fases com instruções curtas (ex.: "planeja a próxima fase", "executa a fase do SETUP") ou pela skill do ritual. O contexto vem dos docs vivos, não do prompt.
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 First seen · 35 lines · 1,127 tokens per session scan A 16adffe58ebf
code-ai-template CLAUDE.md is an instructions file published in the GitHub repository FHDumont/code-ai-template (2 stars, last pushed 22d ago), licensed MIT. It adds 1,127 tokens to every session, about $0.0056 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.