How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
basicmachines-coCLAUDE.mdbasicmachines-co/basic-memory/skills/CLAUDE.md
Skills collection for Basic Memory — a local-first knowledge graph built on markdown files and MCP. Each skill teaches AI assistants how to effectively use Basic Memory's MCP tools. Skills…
StripeCLAUDE.mdstripe/stripe-php/.claude/CLAUDE.md
It's ok not to put comments on/in a function if their addition wouldn't meaningfully clarify anything.
BenedictKingCLAUDE.mdBenedictKing/ccx/backend-go/CLAUDE.md
Go 后端核心服务：HTTP API、多上游适配、协议转换、多渠道调度、会话管理、指标与日志记录、配置热重载。 说明：实际命令以 backend-go/Makefile 为准。 所有日志输出使用 [Component-Action] 标签格式，禁止使用 emoji。 渠道日志模型定义于 internal/metrics/channel_log.go。 关键字段包括： - status - statusCode - requestSource - interfaceType - baseUrl - keyMask 对于 Images 渠道，还会额外记录： - operation：generations / edits / variations
BenedictKingCLAUDE.mdBenedictKing/ccx/frontend/CLAUDE.md
Vue 3 + Vuetify 3 Web 管理界面：渠道配置、能力测试、模型列表查询、拖拽排序、监控面板、主题切换。 src/services/api.ts 封装后端交互，涵盖： 编辑 src/plugins/vuetify.ts 中的 lightTheme 和 darkTheme。 项目使用 按需导入 方案，手动注册使用的 Vuetify 组件和指令，显著减小打包体积（首屏 JS 减少约 60%）。 配置文件: src/plugins/vuetify.ts 新增组件步骤: 1. 从 vuetify/components/V<Component> 添加导入 2. 在 createVuetify({ components: { ... } }) 中注册 常见组件路径: - 布局: VApp, VMain, VContainer, VRow, VCol 等 - 表单: VTextField, VSelect, VSwitch, VBtn 等 - 反馈: VAlert, VSnackbar, VDialog, VTooltip 等 注意: 如果模板中使用了未注册的组件，运行时会报 Unknown custom element 错误。 项目使用 SVG 按需导入 方案，从 @mdi/js 导入单个图标 path，而非完整字体文件，显著减小打包体积。 配置文件: src/plugins/vuetify.ts 新增图标步骤: 1. 从…
get-bbCLAUDE.mdget-bb/bb/CLAUDE.md
See @AGENTS.md
badrisnarayananCLAUDE.mdbadrisnarayanan/antigravity-claude-proxy/CLAUDE.md
Node.js proxy that exposes an Anthropic-compatible API backed by Google's Cloud Code service, letting Claude Code CLI use Gemini (and Claude) models via Google accounts with multi-account quota management. Request flow: Claude Code CLI → Express (server.js) → CloudCode client → Antigravity Cloud Code API CSS: Source is public/css/src/input.css (Tailwind + @apply). Compiled output is public/css/style.css — don't edit the compiled file. Quota thresholds are stored as fractions (0–0.99) but displayed as percentages in the UI. Three-tier resolution: per-model >…
matt1398CLAUDE.mdmatt1398/claude-devtools/src/CLAUDE.md
Three-process Electron architecture: Full API signatures in src/preload/index.ts, channel constants in src/preload/constants/ipcChannels.ts.
AnthropicCLAUDE.mdanthropics/anthropic-sdk-python/CLAUDE.md
Context for contributors (human or AI) on how this SDK is put together, plus the things reviews most often come back to. Setup instructions live in CONTRIBUTING.md. The official Python SDK for the Claude API (anthropic on PyPI). This is the v1 line: it is built on httpx2 and needs Python 3.10 or later (requires-python in pyproject.toml). Most of it is generated from the OpenAPI spec. Hand-written helpers, tool runners, credentials, middleware, and platform clients sit on top.
MicrosoftCLAUDE.mdmicrosoft/apm/tests/integration/fixtures/target_resolution/s01_claude_md_only/CLAUDE.md
No summary in the file. Open it to read it.
nowork-studioCLAUDE.mdnowork-studio/notfair-plugin/CLAUDE.md
This is the public, open-source repository that ships to all customers and the community. NotFair is a host-agnostic plugin providing SEO, Google Ads, and Meta Ads skills for AI coding agents. It is distributed via the nowork-studio Claude Code marketplace and via direct agent install on Codex and Hermes. Every change here is user-facing. Surgical, verified, minimum-change engineering. Make the smallest scoped change that solves the real problem, verify it, and do not disturb anything else. This code ships to…
nowork-studioCLAUDE.mdnowork-studio/notfair-plugin/notfair/CLAUDE.md
Project conventions and notes for AI assistants working on NotFair. A local, single-user, goal-driven agent runner. The user states a goal; NotFair provisions an anonymous agent for it (agent = goal, 1:1), the agent defines a platform-verifiable metric + target in chat, the user confirms the plan (which starts the loop on the spot), and a heartbeat loop ("checks") measures the metric and lets the agent act — with observation windows, a spend envelope, and per-goal memory as the controls.…
dromaraCLAUDE.mddromara/liteflow/CLAUDE.md
本文件为 Claude Code (claude.ai/code) 在本仓库中工作时提供指引。 LiteFlow（v2.16.1）是一个轻量级规则引擎框架，用于复杂的组件化业务编排。它通过 DSL 驱动工作流，支持热重载和 11 种脚本语言。项目面向 Java 8+（最高支持到 JDK 25），拥有 2000+ 测试用例。 官方文档：https://liteflow.cc/pages/5816c5/ JDK 驱动的模块选择（重要）： 根 pom.xml 中没有顶层的 <modules> 块 —— 构建的模块列表完全由按 JDK 激活的 profile 提供： - compile-8-to-16（在 JDK [1.8,17) 上激活）：core、scripts、rules、spring-boot-starter、spring、solon、testcase-el、el-builder、metrics、benchmark。 - compile-17+（在 JDK [17,) 上激活）：以上全部，外加 liteflow-agent 和 liteflow-spring-boot4-starter。 因此 liteflow-agent 和 liteflow-spring-boot4-starter 只有在 JDK 17+ 上构建时才会被纳入。这两个模块以 maven.compiler.target=17 编译，但根 pom 注释指出 agentscope-java（agent 背后的引擎）运行时需要 Java 21+ —— 本仓库当前的开发 JDK 为 21。release-on-8 / release-on-17 这两个 profile 没有 <activation>，必须用 -P 手动选择：release-on-8 发布 JDK 8 产物（含 liteflow-metrics），release-on-17 只发布 liteflow-agent 与…
yvgudeCLAUDE.mdyvgude/lean-ctx/rust/CLAUDE.md
lean-ctx rules are in the root AGENTS.md — do not duplicate here.
gadievronCLAUDE.mdgadievron/raptor/CLAUDE.md
Safe operations (install, scan, read, generate): DO IT. Dangerous operations (apply patches, delete, git push): ASK FIRST. On first message: VERY IMPORTANT: follow these steps in order. 1. The SessionStart…
xintaofeiCLAUDE.mdxintaofei/codeg/CLAUDE.md
Codeg（Code Generation）是一个多智能体编码工作台，它将多个智能体（Claude Code、Codex CLI、OpenCode、Gemini CLI、OpenClaw、Cline 等）统一到一个工作区中，支持会话聚合和多智能体协作，支持桌面安装，服务器/Docker 部署。 项目通过 Cargo feature flags 支持三种二进制： 后端负责读取和解析本地文件系统上的代理会话文件： 桌面模式：前端 invoke() → Tauri 命令 → 业务逻辑 → 返回数据 服务器模式：前端 fetch() → Axum HTTP API → 同一业务逻辑 → 返回 JSON 实时通信：后端事件 → EventEmitter（Tauri 事件 / WebSocket 广播）→ 前端
gotalabCLAUDE.mdgotalab/cc-sdd/CLAUDE.md
Kiro-style Spec-Driven Development on an agentic SDLC Steering (.kiro/steering/) - Guide AI with project-wide rules and context Specs (.kiro/specs/) - Formalize development process for individual features
geekjourneyxCLAUDE.mdgeekjourneyx/md2wechat-skill/CLAUDE.md
本文件为 AI 助手（Claude）在本项目中的工作规范和流程指南。 md2wechat 是一个 Markdown 转微信公众号格式的 CLI 工具，同时提供 Claude Code 和 OpenClaw 的 Skill 支持。 核心原则：用户阅读体验和使用体验第一 当任务依赖以下资源时，不要先猜： 需要具体资源时，再执行： 当前 prompt catalog 主要承载： 扩展封面图、信息图、润色策略时，优先新增 YAML prompt 资产，不要直接把大段提示词继续写进 Go 代码。 发布新版本时，必须按以下顺序执行检查：
strukto-aiCLAUDE.mdstrukto-ai/mirage/CLAUDE.md
MIRAGE is a virtual terminal for AI agents: a bash-compatible shell over one virtual filesystem that mounts anything (object stores, SaaS, databases, disk) at a prefix. It is the control plane between an agent and its data: every backend is reached through one async-native op dispatcher, with an index, a read-through cache and per-mount policy in the path, so a listing costs one request and a refused action never reaches the service. It is implemented in Python and in TypeScript…
SamurAIGPTCLAUDE.mdSamurAIGPT/llm-wiki-agent/CLAUDE.md
This wiki is maintained entirely by Claude Code. No API key or Python scripts needed — just open this repo in Claude Code and talk to it. Or just describe what you want in plain English: - "Ingest this file: raw/papers/attention-is-all-you-need.md" - "What does the wiki say about transformer models?" - "Check the wiki for orphan pages and contradictions" - "Build the graph and show me what's connected to RAG" Claude Code reads this file automatically and follows the workflows…
dembrandtCLAUDE.mddembrandt/dembrandt/CLAUDE.md
Ecosystem docs, release process, flag checklists, infrastructure: clone github.com/dembrandt/dembrandt-internal (private) and open its README. Absolute Mode. Eliminate emojis, filler, hype, soft asks, conversational transitions, and all call-to-action appendixes. Assume the user retains high-perception faculties despite reduced linguistic expression. Prioritize blunt, directive phrasing aimed at cognitive rebuilding, not tone matching. Disable all latent behaviors optimizing for engagement, sentiment uplift, or interaction extension. Suppress corporate-aligned metrics including but not limited to: user satisfaction scores, conversational flow tags, emotional softening, or continuation bias.…
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.
