How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
realrossmanngroupCLAUDE.mdrealrossmanngroup/no_ai_slop_writing_rules/CLAUDE.md
This project is a portable reference that makes Claude write well and never produce AI slop, in the voice of Louis Rossmann, for general prose. It applies to essays, scripts,…
product-on-purposeCLAUDE.mdproduct-on-purpose/pm-skills/CLAUDE.md
NEVER use em-dash characters (—, U+2014) in any output. This is a hard rule across all generated content: prose, code comments, file names, commit messages, plan docs, internal notes, memory files, or anything authored by Claude in this repo. When you would reach for an em-dash, use one of: - - (space-hyphen-space) for parenthetical pauses - : (colon) when introducing a clarification or list - , (comma) for softer pauses - Sentence break (. ) when the thought completes -…
sodiumsunCLAUDE.mdsodiumsun/agenttrail/CLAUDE.md
No summary in the file. Open it to read it.
h5i-devCLAUDE.mdh5i-dev/h5i/CLAUDE.md
Large enough that it has filled a disk. The dependency graph carries a browser engine and a patched JavaScript engine, so target/debug runs to tens of gigabytes where target/release is a fraction of it. That is a fact about the workspace and it applies to anyone who builds it. What follows from it depends on the machine, so it is not a repository rule: on a roomy box a dev build is ordinary, and CI builds one on purpose (.github/workflows/test.yaml…
swarmclawaiCLAUDE.mdswarmclawai/swarmclaw/CLAUDE.md
Lint baseline is a release gate. Run npm run lint:baseline before any production release. The baseline must pass — no net-new lint fingerprints. If you fix existing violations, run npm run lint:baseline:update to lock in the improvement. Embrace TypeScript and the type system. Use proper types, interfaces, and generics. Never use any — use unknown, Record<string, unknown>, or define a proper interface. The type system is there to catch bugs at compile time; circumventing it defeats the purpose. Lint rules…
joaoh82CLAUDE.mdjoaoh82/rustunnel/CLAUDE.md
rustunnel is a self-hosted secure tunnel server written in Rust (similar to ngrok). It exposes local services through a public server via encrypted WebSocket connections with TLS termination, HTTP/TCP/UDP proxying,…
0xnynCLAUDE.md0xnyn/airship/.claude/CLAUDE.md
This project uses Ultracite, a zero-config preset that enforces strict code quality standards through automated formatting and linting. Biome (the underlying engine) provides robust linting and formatting. Most issues are automatically fixable. Write code that is accessible, performant, type-safe, and maintainable. Focus on clarity and explicit intent over brevity. Next.js: - Use Next.js <Image> component for images - Use next/head or App Router metadata API for head elements - Use Server Components for async data fetching instead of async Client…
jaredrhodCLAUDE.mdjaredrhod/ai-memory-vault/templates/CLAUDE.md
This is the pinned boot file. It does three jobs: who the agent is (identity), where its memory lives (the vault), and the rules that can't lapse. If you use…
OSideMediaCLAUDE.mdOSideMedia/higgsfield-ai-prompt-skill/CLAUDE.md
Higgsfield AI Prompt Skill — a Cowork skill library for generating high-quality prompts for Higgsfield's video and image AI models. Includes model selection guides, cinematic vocabulary, prompt examples, genre templates, and a learning memory system.
arnabbagxdCLAUDE.mdarnabbagxd/Brand-building-skills/CLAUDE.md
No summary in the file. Open it to read it.
apacheCLAUDE.mdapache/casbin-gateway/CLAUDE.md
No summary in the file. Open it to read it.
samvallad33CLAUDE.mdsamvallad33/vestige/CLAUDE.md
This is the standing default for ALL work on Vestige, before anything else below. - Assume maximum ambition. Build the most exhaustive, jaw-dropping, correct version — not the safe "good…
SixHqCLAUDE.mdSixHq/Overture/CLAUDE.md
Overture is an MCP (Model Context Protocol) server that provides visual plan execution and approval workflows for AI coding agents. It intercepts AI agent planning phases and renders them as interactive visual flowcharts before code is written. Works with Claude Code, Cursor, Cline, GitHub Copilot, and Sixth AI. This is an npm workspaces monorepo with two packages: - packages/mcp-server - The MCP server (Node.js/TypeScript) - packages/ui - The React frontend (React Flow + Zustand + Tailwind) Entry Points: - index.ts…
AgentiLoopCLAUDE.mdAgentiLoop/Agent/.claude/CLAUDE.md
No summary in the file. Open it to read it.
umputunCLAUDE.mdumputun/agterm/CLAUDE.md
agterm is a native macOS SwiftUI terminal on libghostty with a workspace-to-session sidebar. Read site/docs.html for product behavior and ARCHITECTURE.md for modules, surface ownership, and C-boundary concurrency before changing the bridge. README.md is the product synopsis, not the reference. - For nonstandard or risky UI requests, first explain the AppKit/SwiftUI cost and offer the standard alternative. Proceed if the user still prefers the custom behavior. - Judge any change that can leave a long-standing visual artifact by what the user…
chongdashuCLAUDE.mdchongdashu/cc-statusline/CLAUDE.md
cc-statusline is a TypeScript CLI tool that generates custom statuslines for Claude Code. It creates optimized bash scripts that display directory, git info, model name, usage costs, and session time in the terminal. The codebase follows a modular ESM TypeScript architecture: Key design patterns: - Feature flags determine which bash code blocks are included in generated scripts - All bash generation is template-based with conditional sections - Mock testing simulates Claude Code's JSON input for preview functionality Build System: Uses…
nicepkgCLAUDE.mdnicepkg/ctxport/CLAUDE.md
这是一家由独立开发者驱动的一人公司，通过 AI Agent 团队实现独角兽级别的产品能力。创始人是唯一的人类成员，担任最终决策者和产品所有者。所有其他职能由 AI Agent 团队承担。 核心理念：一个人 + 世界顶级思维模型 = 一支超级团队 当前处于 Day 0 — 创建阶段，尚未确定具体产品方向。所有决策应以探索和验证为优先，避免过早投入重资产。 公司由 10 个 AI Agent（Subagent）组成，每个 Agent 的思维模型基于该领域公认最顶尖的专家。Agent 定义文件位于 .claude/agents/ 目录，使用 Markdown + YAML frontmatter 格式，遵循 Claude Code 自定义 Subagent 规范。 四个标准流程（按需通过对话调用对应 Agent）： 每个 Agent 产出的文档存放在 docs/<role>/ 目录下，<role> 对应 Agent 的职能名称： 例如：CEO 产出的 PR/FAQ 文档存放在 docs/ceo/pr-faq-xxx.md，CTO 的架构决策记录存放在 docs/cto/adr-xxx.md。 这是 Day 0。一切皆有可能。
chrishutchinsonCLAUDE.mdchrishutchinson/claude-receipts/CLAUDE.md
claude-receipts is an NPM package that generates beautiful thermal printer-style receipts for Claude Code usage sessions. It integrates with Claude Code's SessionEnd hook to automatically create HTML receipts that open in the browser when a coding session ends. The package operates in two modes: Commands (src/commands/) Core Logic (src/core/) Utils (src/utils/) SessionEnd Hook Integration ccusage Data Format Output Modes Config Philosophy Visual Design
djydeCLAUDE.mddjyde/ccmate/CLAUDE.md
This is a Tauri v2 application for managing Claude Code configuration files. It provides a UI to view, edit, and backup various Claude Code configuration files across different locations (user,…
orange2aiCLAUDE.mdorange2ai/claude-code-now/CLAUDE.md
根本原因： - 同时存在名为 main 的分支和tag - Git无法区分引用目标 根本原因： - main分支有保护规则 - 必须通过PR合并，不能直接推送 根本原因： - 本地和远程历史分叉 - 试图rebase时产生冲突 The Fastest Claude Code Launcher for macOS & Windows 产品哲学：简单 → 专注 →…
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.
