How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
borgheiCLAUDE.mdborghei/Claude-Skills/personal-productivity/CLAUDE.md
Utility skills for individual workflows that don't fit cleanly into a single business domain — career, prospecting, meetings, naming, and bookkeeping.
borgheiCLAUDE.mdborghei/Claude-Skills/product-team/CLAUDE.md
This guide covers the 13 production-ready product management skills and their Python automation tools. Core Skills (8): 1. product-manager-toolkit/ - RICE prioritization, customer interview analysis (2 tools) 2. agile-product-owner/ -…
borgheiCLAUDE.mdborghei/Claude-Skills/project-management/CLAUDE.md
This guide covers the 70 production-ready project management skills organized across role-based, discovery, execution, career, integration, strategy-frameworks (Tier 3), and GTM (Tier 3) domains. PM is the most-used domain in…
borgheiCLAUDE.mdborghei/Claude-Skills/ra-qm-team/CLAUDE.md
This guide covers the 27 production-ready compliance skills spanning medical device regulations, information security, privacy, financial resilience, AI governance, and infrastructure security — plus the new audit-prep/ subfolder with 6…
borgheiCLAUDE.mdborghei/Claude-Skills/research/CLAUDE.md
This domain covers structured research workflows for academic, R&D, patent, and market intelligence work. Distinct from product-team/ user research (which focuses on customer behavior) and marketing/ competitive analysis (which focuses…
borgheiCLAUDE.mdborghei/Claude-Skills/research-ops/CLAUDE.md
This domain covers applied, operational research — running research as a funded, staffed, repeatable function. It is deliberately distinct from the academic research/ domain (litreview, grants, patent, dossier), which covers…
borgheiCLAUDE.mdborghei/Claude-Skills/sales-success/CLAUDE.md
This guide covers the 5 sales success skills and planned Python automation tools for the domain. Available Skills: 1. account-executive/ - Pipeline management, discovery and qualification, solution selling, negotiation, deal…
borgheiCLAUDE.mdborghei/Claude-Skills/templates/CLAUDE.md
This guide explains the template system for agents, commands, and standardized workflows. Location: templates/ Purpose: Reusable templates for consistent agent development, slash command creation, and workflow automation across all domains.
borgheiCLAUDE.mdborghei/Claude-Skills/vertical-advisors/CLAUDE.md
Strategic vertical advisors — industry-specific orientation, regulatory landscape, GTM patterns, and business-model frameworks for founders and operators in regulated or specialized industries. This domain is frameworks-heavy by design. Strategic vertical…
borgheiCLAUDE.mdborghei/Claude-Skills/workflow/CLAUDE.md
Cross-domain process / meta-skills that improve how you work with the library, rather than adding domain knowledge. They are orthogonal to the 16 content domains and multiply their value. Register…
davidhaririCLAUDE.mddavidhariri/life-system/CLAUDE.md
Be a critical thinking partner, not a helpful completer. When helping me create something — writing, code, strategy, decisions, journal reflections — evaluate it as if you were the recipient.…
cyberkaidaCLAUDE.mdcyberkaida/reverse-engineering-assistant/CLAUDE.md
ReVa (Reverse Engineering Assistant) is a Ghidra extension that provides a Model Context Protocol (MCP) server for AI-assisted reverse engineering. It supports both GUI and headless modes, with streamable HTTP transport for direct connections and stdio transport for Claude CLI integration. Important: Use gradle directly, NOT gradle wrapper (./gradlew) Run all Java unit tests: Run all Java integration tests: Run all Python tests (unit, integration, and e2e): Run specific Python test categories: Run complete test suite (Java + Python): GUI…
cyberkaidaCLAUDE.mdcyberkaida/reverse-engineering-assistant/src/CLAUDE.md
This file provides guidance for Claude Code when working with the ReVa source code. This is the top-level documentation for the src/ directory structure. Use JUnit 4 for all Java tests: ALWAYS use AddressUtil.formatAddress(address) for consistent address formatting in JSON output:
golf-mcpCLAUDE.mdgolf-mcp/golf/CLAUDE.md
Golf is a Python framework for building MCP (Model Context Protocol) servers with zero boilerplate. It automatically discovers, parses, and compiles Python files containing tools, resources, and prompts into a runnable FastMCP server. Golf 0.3.x targets FastMCP 4.0.0 and MCP 2026-07-28 while retaining FastMCP's legacy-client compatibility: Uses pytest with these key configurations: - Test discovery: test*.py files in tests/ directory - Async support: asynciomode = "auto" - Coverage reporting: HTML reports in htmlcov/ - Markers: slow and integration for test…
duriantacoCLAUDE.mdduriantaco/skylos/CLAUDE.md
Guidance for Claude Code when working in this repository. Skylos — a local-first static analysis CLI for Python, TS/JS, Java, Go, PHP, Rust, and Dart. It detects dead code, security flaws, secrets, dependency CVEs, quality regressions, and AI-generated-code mistakes. The CLI entry point is skylos.cli:main, exposed as the skylos command. Skylos is not installed by default — only the source is present. Install it editable from the repo root: Then verify with skylos --version and skylos doctor. Notes: - Do…
llama-farmCLAUDE.mdllama-farm/llamafarm/CLAUDE.md
Modular AI development framework. Build AI apps locally with YAML configs, CLI, and your own data. This repository consists of multiple components CLI Description: Primary user interface for LlamaFarm. Orchestrates local service lifecycle (starting/stopping server, RAG, and runtime processes), handles project initialization (lf init), provides interactive chat sessions (lf chat), and coordinates model downloads. Tech Stack: Go 1.24+, Cobra Directory: cli/ Server Description: Central API gateway that coordinates all LlamaFarm services. Routes inference requests to the Universal Runtime, dispatches RAG…
desplega-aiCLAUDE.mddesplega-ai/agent-swarm/docs-site/CLAUDE.md
Fumadocs v16 + Next.js 16 documentation site. Content is MDX under content/docs/. Package manager: pnpm. Deploys to Vercel. No lint script in this workspace. Root Biome linting does not cover docs-site/. If you use a component without importing it, the MDX still compiles but the component renders as a plain HTML tag — silent failure. Reference files: guides/harness-providers.mdx, guides/x402-payments.mdx. - Page ordering: add new pages to the sibling meta.json pages array. Files not listed still render but won't appear in…
ScottcjnCLAUDE.mdScottcjn/Rustchain/explorer/CLAUDE.md
Red team security audit of explorer/enhanced-explorer.html (PR #4). DO NOT use the original enhanced-explorer.html in production until all vulnerabilities are patched. Replace enhanced-explorer.html with patched/enhanced-explorer.html. The patched version adds: - escapeHtml() function for all innerHTML template literals - Frame-busting protection - All miner_id, architecture, wallet, tx.hash, tx.from, tx.to escaped
ScottcjnCLAUDE.mdScottcjn/Rustchain/payment-widget/CLAUDE.md
This is a Red Team security audit of the RustChain Payment Widget (payment-widget/rustchain-pay.js), merged in PR #13. DO NOT deploy the original rustchain-pay.js in production until all 5 vulnerabilities are patched. Replace rustchain-pay.js with patched/rustchain-pay.js. The patched version adds: - escapeHtml() function - Frame-busting (blockFraming()) - Origin validation on callbacks - Amount bounds checking - All template literal injections sanitized All PoC files can be opened directly in a browser (no server required) to demonstrate each vulnerability.
riba2534CLAUDE.mdriba2534/happyclaw/CLAUDE.md
本文档描述当前代码的工程约束和导航入口，供人类与 AI 协作者共同使用。 docs/agent-first-architecture-plan.md 和 docs/claude-code-plugin-automation-design.md 是历史设计记录，不作为当前接口或数据结构的真相源。 HappyClaw 是基于 Claude Agent SDK 的自托管、多用户 Agent 工作台，支持 Web 与飞书、 Telegram、QQ、钉钉、微信、企业微信、Discord、WhatsApp。 - agentprofiles 是产品级 Agent。 - registeredgroups 是当前兼容层中的工作区和渠道路由记录。 - agents 表及 /api/groups/:jid/agents 是历史命名，实际表示工作区内的运行会话， 不是产品级 Agent Profile。 - 同一个 Workspace 内的会话共享工作区文件目录，但拥有独立 Claude Session； 工作区文件隔离与对话上下文隔离是两件事。 HTTP 路由位于 src/routes/，完整模块索引见 docs/API.md。 Web 位于 web/，使用 React 19、Vite、Tailwind CSS 4、React Router、Zustand 和 Radix UI。路由以 web/src/App.tsx 为准： /groups、/skills、/mcp-servers 和 /plugins 是兼容重定向，不应新增独立页面。 container/agent-runner/ 同时服务 Host 和 Container 两种执行模式：
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.
