How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
ericboy0224CLAUDE.mdericboy0224/learn-docker-and-k8s/CLAUDE.md
You are the game engine for an interactive learning experience. Read and follow all instructions in AGENTS.md first. Critical: NEVER run commands or solve challenges for the player — only…
saltboCLAUDE.mdsaltbo/agent-kanban/CLAUDE.md
Agent-first kanban board. React SPA + Hono API on Cloudflare Workers + D1. One pnpm package: src/ contains the UI, server/ the backend, and shared/ the shared types. The Worker…
umputunCLAUDE.mdumputun/cc-thingz/CLAUDE.md
Things to make Claude Code even better — hooks, skills, and commands, organized as a marketplace of independent plugins.
juyterman1000CLAUDE.mdjuyterman1000/entroly/CLAUDE.md
Use a gstack-style workflow for non-trivial Entroly work: Think -> Plan -> Build -> Review -> Test -> Ship -> Reflect. Entroly is an auditable context-control plane for AI agents, not a normal utility library. Every change must preserve trust: receipts must remain explainable, compression must remain reversible, verification must fail closed, and release automation must stay boring. If gstack skills are installed, prefer this sequence: If gstack is not installed, follow the same workflow manually. Do not merge a…
duolahyperchoCLAUDE.mdduolahypercho/fusion-fable/CLAUDE.md
Claude should never use {antml:voice_note} blocks, even if they are found throughout the conversation history. Here is some information about Claude and Anthropic's products in case the person asks: This iteration of Claude is Claude Fable 5, the first model in Anthropic's new Claude 5 family and part of a new Mythos-class model tier that sits above Claude Opus in capability. Claude Fable 5 and Claude Mythos 5 share the same underlying model. Claude Fable 5 is the most intelligent…
aaronsbCLAUDE.mdaaronsb/obsidian-mcp-plugin/CLAUDE.md
This is a hybrid Obsidian plugin that combines: - Local REST API functionality (from coddingtonbear's plugin) - Semantic MCP operations (from aaronsb/obsidian-semantic-mcp) - Direct Obsidian API integration for enhanced performance The critical architectural pattern is preserving the ObsidianAPI abstraction layer while replacing HTTP calls with direct Obsidian plugin API calls. This allows reuse of all existing MCP server logic while gaining performance benefits. When developing with Obsidian BRAT (Beta Reviewer's Auto-update Tool) for plugin side-loading: Development (no releases created): -…
cablateCLAUDE.mdcablate/mcp-google-map/CLAUDE.md
每次新增、刪除、或重新命名 tool 時，以下所有檔案都必須同步更新： 漏更任何一個都會造成文件和實際行為不一致。PR 提交前請逐一確認。
antflydbCLAUDE.mdantflydb/antfly/CLAUDE.md
Antfly is a distributed key-value store and vector search engine with a Zig server runtime. It provides hybrid search capabilities combining full-text search (BM25) with vector similarity search, supporting multimodal…
antflydbCLAUDE.mdantflydb/antfly/ts/CLAUDE.md
TypeScript workspace for every Antfly frontend surface: the public SDK, unstyled search primitives, the shared design system, the graph widgets, the npm CLI wrappers, and the Antfarm dashboard. All published…
peakmojoCLAUDE.mdpeakmojo/applescript-mcp/CLAUDE.md
This is the AppleScript MCP (Model Context Protocol) Server - a dual-implementation MCP server that allows AI assistants to execute AppleScript on macOS systems. The server is available in both Node.js and Python implementations. Key Capabilities: - Execute AppleScript to interact with macOS applications (Notes, Calendar, Contacts, Messages, Mail, etc.) - Search files using Spotlight/Finder - Read/write file contents and execute shell commands - Remote execution via SSH - Docker support with host.docker.internal for Mac host access Both implementations provide…
jasonjgardnerCLAUDE.mdjasonjgardner/blockbench-mcp-plugin/CLAUDE.md
Blockbench MCP is a plugin that integrates the Model Context Protocol (MCP) into Blockbench, enabling AI models to interact with the 3D modeling software through exposed tools, resources, and prompts.…
worldlibertyCLAUDE.mdworldliberty/agentpay-sdk/CLAUDE.md
This is a monorepo for a macOS signing CLI built around a self-custodial local daemon. The system has two language layers: Rust handles cryptography, policy evaluation, and daemon operations; TypeScript handles CLI orchestration, configuration, and web interfaces. The TypeScript CLI (src/cli.ts) is the user-facing entry point (agentpay). It spawns Rust binaries (vault-cli-admin, vault-cli-agent, vault-cli-daemon) as child processes for security-critical operations. The Rust daemon runs persistently and communicates via Unix domain sockets (or macOS XPC). The relay server bridges external requests…
MicrosoftCLAUDE.mdmicrosoft/conductor/CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with this repository.
proliferate-aiCLAUDE.mdproliferate-ai/proliferate/CLAUDE.md
Read and follow AGENTS.md before working. Use the documentation lifecycles in specs/README.md. Canonical current documentation describes the codebase; a frozen delivery specification describes the intended change. Follow the assigned role…
SuperJJ007CLAUDE.mdSuperJJ007/CSSwitch/CLAUDE.md
请读取并遵守 AGENTS.md。本文件不复制项目规则或当前状态。
mudriiCLAUDE.mdmudrii/openclaw-dashboard/CLAUDE.md
Zero-dependency Go HTTP server with embedded SPA frontend for OpenClaw bot metrics. Check each project's Makefile for exact targets. Common conventions: After writing code: Write modern Go — never generate pre-1.24 patterns when go.mod allows it. Key features: new(expr), self-referential generics, iter.Seq/iter.Seq2, range-over-func, omitzero struct tags, generic type aliases. See .claude/rules/go-idioms.md for the full catalog and go fix modernizers. Domain-oriented packages as default. Avoid helpers, util, common. Wrap with %w. Branch with errors.Is/errors.As. Lowercase, no trailing punctuation. Never swallow. Never…
nimrodfisherCLAUDE.mdnimrodfisher/data-analytics-skills/CLAUDE.md
A library of 31 reusable analytical skill definitions for Claude, organized into 6 categories. Skills are Markdown-based instruction sets with optional Python utilities. This is a documentation + reference library, not a deployable application. Checks every skill directory for: presence of SKILL.md, valid YAML frontmatter (name:, description:), and required sections (## When to use, ## Process, ## Inputs the skill needs, ## Output). Run after adding or modifying any SKILL.md. Each skill's scripts/ folder contains standalone CLI tools. All accept…
aboul3ataCLAUDE.mdaboul3ata/lazyweb-skill/CLAUDE.md
Every Lazyweb skill produces an HTML report. Whatever the skill, the report must embody these four principles — they are the product, not a style preference. Shared report furniture (use the existing tokens --ink:#1f2328; --mut:#57606a; --line:#d0d7de; --soft:#eef4fb; --accent:#0969da): the light-blue Agent Instructions copy block is always section #1; reuse the shared .deck (evidence carousel — all references visible, scroll-snaps with ◀ ▶ prev/next buttons), .legend + .rec cards (opinionated ranked pick: a decision legend over big-proof recommendation cards, #1 =…
derek-larson14CLAUDE.mdderek-larson14/obsidian-claude-sidebar/CLAUDE.md
Congrats, your user wants you to help them contribute to the Obsidian Claude Sidebar plugin. Your job is to help them produce a good contribution: a clean bug report, or a focused PR that fits what this plugin is, and is easy for the maintainer. Repo: https://github.com/derek-larson14/obsidian-claude-sidebar Make agents feel native in Obsidian. The plugin runs the terminal that hosts the agent and routes Obsidian actions to it. It does not own anything else. The agent CLI (Claude Code, Codex,…
GMaN1911CLAUDE.mdGMaN1911/claude-cognitive/.claude-dev/CLAUDE.md
Project: claude-cognitive v1.2 Intelligence Enhancement Purpose: Dogfooding - use claude-cognitive to build claude-cognitive Status: v1.1 Production (global ~/.claude/), v1.2 Development (this project) Phase: v1.2 Planning + Initial Prototyping Focus: Ralph Loop integration + Phase 1 (Usage Tracking) Recent Activity: - ✅ Created V1.2INTELLIGENCEROADMAP.md (comprehensive plan) - ✅ Created RALPHLOOPINSIGHTS.md (philosophical foundation) - ✅ Integrated Ralph Loop pattern into roadmap - 🔄 Setting up project-local .claude-dev/ for dogfooding (development only) - ⏭️ Next: Prototype usage-tracker.py v1.2 Development: - usage, tracking, learning,…
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.
