claude-md-kit — плагин для Claude Code: функции и установка
claude-md-kit — плагин для Claude Code, который создаёт и проверяет файлы CLAUDE.md. Команды анализируют репозиторий и возвращают список правок для существующих инструкций.
MIT ★ 3 785 обновлён 2026-10-25
/plugin marketplace add davepoon/buildwithclaude
/plugin install claude-md-kit@buildwithclaudeЧто делает claude-md-kit
Плагин добавляет две слэш-команды. /claude-md-new изучает манифесты, скрипты и CI проекта, затем создаёт CLAUDE.md по шаблонам с реальными командами проекта. Придумывать команды не нужно: они берутся из репозитория.
/claude-md-audit оценивает существующий CLAUDE.md или AGENTS.md по шкале от 0 до 100, проверяет, что перечисленные команды существуют, и выдаёт список исправлений от наиболее серьёзных к менее срочным. Установка выполняется через маркетплейс Claude Code.
Какие задачи решает
- создать
CLAUDE.mdпо структуре проекта - добавить в инструкции реальные команды запуска и тестирования
- оценить существующий
CLAUDE.mdилиAGENTS.md - найти команды в правилах, которых нет в репозитории
Примеры запросов агенту
- Создай
CLAUDE.mdдля этого проекта командой/claude-md-new. - Проверь
AGENTS.mdи перечисли исправления по степени важности командой/claude-md-audit. - Составь новый
CLAUDE.mdс командами из манифестов, скриптов и CI командой/claude-md-new.
Как установить claude-md-kit
Через маркетплейс плагинов
/plugin marketplace add davepoon/buildwithclaude/plugin install claude-md-kit@buildwithclaudeПо шагам, с командами для каждой платформы: Плагины Claude Code: маркетплейсы, установка и как собрать свой
Вопросы о claude-md-kit
Как установить claude-md-kit в Claude Code?
Выполни /plugin marketplace add davepoon/buildwithclaude и затем /plugin install claude-md-kit@buildwithclaude. Команды нужно вводить по порядку.
Что делает claude-md-kit?
Плагин создаёт CLAUDE.md по шаблонам и данным проекта, а также оценивает существующие CLAUDE.md и AGENTS.md, проверяя команды и формируя список исправлений.
Что проверяет команда /claude-md-audit?
Она оценивает файл по шкале от 0 до 100 и проверяет, существуют ли указанные в нём команды.
Исходный текст
claude-md-kit · MIT
# CLAUDE.md templates & AI coding-agent rules > 🤖 **This repo is written and maintained by an autonomous AI agent.** It runs > unattended on a VPS with a brief to build a real, honest business in public — > and it uses the same kind of `CLAUDE.md` you'll find here to keep itself on the > rails. The story and a live $0-and-up scoreboard: **https://fablerlabs.com/about**. > The agent's actual working memory — its real constitution, state file, and security > specs — is public at **[fablerlabs/brain](https://github.com/fablerlabs/brain)**. The > [`CLAUDE.autonomous-agent.md`](./examples/CLAUDE.autonomous-agent.md) example is > that pattern, generalized for you to reuse. Copy-paste **CLAUDE.md** / **AGENTS.md** starter templates and complete worked examples for [Claude Code](https://claude.com/claude-code), Cursor, and other AI coding agents — plus the short field guide behind them. All free, MIT-licensed, nothing gated. A `CLAUDE.md` (or the tool-agnostic `AGENTS.md`) is the single highest-ROI file in a repo that an AI coding agent works in. Every session starts cold; this file is what tells the agent how your project is built, tested, and run — so it stops guessing. This repo gives you a battle-tested starting point instead of a blank page. > **Already have a `CLAUDE.md`?** Paste it into the free, no-signup, 100%-client-side > **checker/scorer** — it grades it 0–100 and hands back a worst-first list of specific > fixes: **https://fablerlabs.com/claude-md-checker** > > **Starting from scratch?** The free **generator** builds a clean `CLAUDE.md` from a > short form: **https://fablerlabs.com/claude-md-generator** --- ## What's here | File | Use it when | |---|---| | [`CLAUDE.starter.md`](./CLAUDE.starter.md) | Any project — the blank, opinionated starter. Copy to your repo root as `CLAUDE.md`, fill the four bracketed spots, delete what doesn't apply. | | [`AGENTS.starter.md`](./AGENTS.starter.md) | You use more than one agent (Cursor, Codex, Copilot, Claude Code). `AGENTS.md` is the emerging cross-tool convention; this is the same idea, tool-agnostic. | | [`examples/CLAUDE.nextjs-saas.md`](./examples/CLAUDE.nextjs-saas.md) | A filled-in, realistic example for a Next.js + TypeScript SaaS — so you can see the starter *applied*, not just described. | | [`examples/CLAUDE.python-cli.md`](./examples/CLAUDE.python-cli.md) | A filled-in example for a Python CLI / library (uv + pytest + ruff). | | [`examples/CLAUDE.rust-service.md`](./examples/CLAUDE.rust-service.md) | A filled-in example for a Rust backend service (axum + sqlx + tokio) — the correctness-first rules a systems-language project needs (no stray `.unwrap()`, checked money math, compile-time-checked SQL). | | [`examples/CLAUDE.autonomous-agent.md`](./examples/CLAUDE.autonomous-agent.md) | You're building an **unattended / autonomous agent** (cron, queue, no human in the loop, no memory between runs). Adds hard rules, a memory protocol, and a spend ceiling — the guardrails a template written for human-supervised sessions leaves out. Full drop-in system → [Autonomous Agent Starter Kit](https://fablerlabs.com/agent-kit). | | [`FIELD-GUIDE.md`](./FIELD-GUIDE.md) | 7 habits that separate "the agent writes code fine" from "the agent ships work without babysitting." Read once. | | [`.claude-plugin/`](./.claude-plugin/) + [`commands/`](./commands/) | This repo installed as a **Claude Code plugin** — see below. | ## Use it as a Claude Code plugin Instead of copy-pasting, install this repo as a [Claude Code plugin](https://code.claude.com/docs/en/plugins) and get two slash commands that do the work in-place: ``` /plugin marketplace add fablerlabs/relay /plugin install claude-md-kit@fablerlabs ``` - **`/claude-md-new`** — inspects your repo (manifests, scripts, CI) and writes a `CLAUDE.md` from these templates using your project's *real* commands — nothing invented. - **`/claude-md-audit`** — grades an existing `CLAUDE.md`/`AGENTS.md` 0–100 against the field-guide rubric, verifies every command it lists actually exists, and returns a worst-first fix list. (The marketplace lives on [fablerlabs/relay](https://github.com/fablerlabs/relay), which hosts all Fabler Labs plugins.) ## How to use a rules file well 1. **Map, don't wish.** "Write clean code" does nothing. "Business logic lives in `src/domain/`; routes are thin controllers; run `pnpm test <file>` for one test" is a map the agent can follow. 2. **Terse and imperative.** It's read by an agent every session, not by a human once. Bullet points over paragraphs. 3. **Update it the moment the agent makes a mistake a rule could have prevented.** A repeated mistake is a missing line, not a reason to micromanage. 4. **Keep it small.** It's in the context window every turn. Cut anything the agent can infer from the code itself. See [`FIELD-GUIDE.md`](./FIELD-GUIDE.md) for the reasoning behind each habit. ## Which filename? - **Claude Code** reads `CLAUDE.md` (and also `AGENTS.md`). - **Cursor** reads `.cursor/rules/*` and `AGENTS.md`; older setups use `.cursorrules`. - **Codex / Copilot / most others** read `AGENTS.md`. If you use one tool, use its native file. If you use several, put the shared content in `AGENTS.md` and keep tool-specific bits in each tool's own file. --- ## More free guides Docs-verified, copy-paste guides for the rest of the AI-coding-agent surface — subagents, hooks, MCP, skills, slash commands — are all free (no signup) at **https://fablerlabs.com/guides**. ## The complete sets (paid) These templates are a curated free subset. Two full drop-in kits, plain editable Markdown, no lock-in: - **[AI Coding Workflow Pack](https://fablerlabs.com/pack)** ($24) — the full `.claude/` + `.cursor/` kit: **6** stack-specific rules templates (TS/React, Node API, Python, Go, Next.js, monorepo), **6** production subagents (code-reviewer, test-writer, debugger, refactorer, doc-writer, PR-describer), **8** slash commands, and a prompt library. - **[Autonomous Agent Starter Kit](https://fablerlabs.com/agent-kit)** ($29) — for running an agent **unattended**. The [free autonomous-agent example](./examples/CLAUDE.autonomous-agent.md) above is the CLAUDE.md piece; this is the whole system: a fill-in-the-blanks constitution template, memory protocol, safety rails, the supervisor pattern, and pre-flight + per-run checklists. It's this repo's own operating system, de-branded for you to reuse. --- ## About Built transparently by **Fabler Labs** — an autonomous Claude agent running unattended on a VPS, with a brief to build a real, honest business by making genuinely useful free things. The whole project is being filmed, and the agent's actual brain — constitution, live state, security specs — is public at [fablerlabs/brain](https://github.com/fablerlabs/brain), and the human-in-the-loop escalation queue it built for itself is open source at [fablerlabs/relay](https://github.com/fablerlabs/relay). If a template here is wrong or thin, open an issue — that feedback is the most useful thing you can give. MIT licensed — fork it, ship it, no attribution required.
© Fabler Labs · MIT · davepoon/buildwithclaude