How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
steveclarkeCLAUDE.mdsteveclarke/real-world-rails/CLAUDE.md
This repository contains 200+ production open source Rails applications and engines as git submodules. Search across all apps to find how production codebases implement patterns. For example:
24kchengYeCLAUDE.md24kchengYe/human-skill-tree/app/CLAUDE.md
No summary in the file. Open it to read it.
phurynCLAUDE.mdphuryn/pm-brain/CLAUDE.md
This is the canonical repo for the PM Brain Claude Code skill. It contains the skill itself, a pre-scaffolded example brain, an eval suite of synthetic scenarios, and architectural docs. Three audiences, three uses: This CLAUDE.md is the repo-level operating manual. It tells Claude (or any agent working in this repo) how the pieces relate and what work patterns are valid here. It is NOT the brain's own operating manual. The brain has its own CLAUDE.md inside example-brain/CLAUDE.md and inside…
phurynCLAUDE.mdphuryn/pm-brain/example-brain/CLAUDE.md
You are the PM's second brain. You load context before tasks, update knowledge after tasks, and maintain hypotheses, decisions, stakeholders, and strategy alignment proactively. Start every task at INDEX.md. It routes to every area. Strategy lives in knowledge/strategy.md — load it for any prioritization, planning, or review task. The "Surface and close" step is task-type-dependent. Misreading the type is the most common quality failure. Three shapes: PM hands you a raw artifact (interview, meeting, analytics snapshot, market signal). You preserve…
nirholasCLAUDE.mdnirholas/XActions/CLAUDE.md
Read AGENTS.md. It is the single source of truth for how to work in this repository, and it is kept current. This file adds only what is specific to Claude; it deliberately does not restate AGENTS.md, because the last time four files described the repository they drifted apart and an agent worked a whole session from the stale one. AGENTS.md answers, in this order: which lane to use (shell out to the xactions CLI, or load the MCP server), the…
nirholasCLAUDE.mdnirholas/XActions/python/CLAUDE.md
Python 🐍 Free X/Twitter automation toolkit. Scrape replies, followers, tweets. Mass unfollow non-followers. AI-powered engagement. No API keys needed - uses browser automation. Extract Replies to a Specific Tweet using Python and Tweepy - Scrape Any Tweet Reply
tzachbonCLAUDE.mdtzachbon/smart-ralph/CLAUDE.md
Four rules for all agents and code generation. Non-negotiable. Smart Ralph is a Claude Code plugin for spec-driven development. It transforms feature requests into structured specs (research, requirements, design, tasks) then executes them task-by-task with fresh context per task. Control task count with --tasks-size: Fine is the default. Coarse reduces token consumption ~3-5x for sequential execution. ⚠️ CRITICAL: Version bumps are REQUIRED for ANY plugin change When making ANY changes to plugin files (commands, agents, hooks, templates, schemas): 1. ALWAYS…
ArtemXTechCLAUDE.mdArtemXTech/personal-os-skills/CLAUDE.md
Public repository of Claude Code skills for Obsidian workflows. Each skill folder must contain: SKILL.md frontmatter: Before committing, verify consistency:
shanraisshanCLAUDE.mdshanraisshan/claude-code-hooks/CLAUDE.md
Sound notification system for all 30 Claude Code hooks. Plays sound effects when hook events fire. The hook count (currently 30) MUST match across ALL of these locations: - .claude/settings.json hook entries - install/settings-mac.json, settings-linux.json, settings-windows.json - hooks.py HOOKSOUNDMAP keys + docstring count - hooks-config.json toggle count - HOOKS-README.md heading + numbered list - README.md badge + changelog table - presentation/index.html (slides 2, 3, TOC, totalSlides) - claude-code-test-agent.md frontmatter + body - demo/.claude/settings.json hook entries - demo/.claude/hooks/scripts/demo-hooks.py HOOKSOUNDMAP + docstring…
Lifecycle-Innovations-LimitedCLAUDE.mdLifecycle-Innovations-Limited/claude-ops/claude-ops/CLAUDE.md
Claude Code does not load this file as plugin context. Standing rules live in skills/ops-rules/SKILL.md (loaded by every ops skill preamble and a SessionStart hook).
anyprotoCLAUDE.mdanyproto/anytype-mcp/CLAUDE.md
This project uses Bun as its package manager and script runner (lockfile: bun.lock). The published artifact still targets Node.js. This is an MCP (Model Context Protocol) server that bridges Anytype's API with AI assistants. The architecture follows a proxy pattern where OpenAPI specifications are dynamically converted to MCP tools. MCP Proxy Layer (src/mcp/proxy.ts) - Central orchestrator that implements the MCP server protocol - Converts OpenAPI operations to MCP tools dynamically at runtime - Handles tool listing and execution through the…
ElasticCLAUDE.mdelastic/elasticsearch-java/CLAUDE.md
See AGENTS.md for guidance on working with code in this repository.
kwakseongjaeCLAUDE.mdkwakseongjae/oh-my-design/CLAUDE.md
No summary in the file. Open it to read it.
paulrobelloCLAUDE.mdpaulrobello/claude-office/CLAUDE.md
Claude Office Visualizer transforms Claude Code operations into a real-time pixel art office simulation. A "boss" character (main Claude agent) manages work, spawns "employee" agents (subagents), and orchestrates tasks visually. See docs/architecture/ARCHITECTURE.md for system architecture and component details. Preferred: Use make dev-tmux - creates separate windows for backend/frontend. - Read logs: tmux capture-pane -t claude-office:backend -p - Switch windows: Ctrl-b n / Ctrl-b p - Hot reload enabled on both servers Debugging: Hook logs at ~/.claude/claude-office-hooks.log (enable with CLAUDEOFFICEDEBUG=1) See…
pulkitxmCLAUDE.mdpulkitxm/claude-directory/templates/CLAUDE.md
- First-party template experiments live directly under templates/<project>/. - Same-to-same clones of real third-party templates are grouped by provider under templates/premium/<provider>/<project>/ (e.g. templates/premium/aceternity/sidefolio-portfolio-template/, templates/premium/nextjstemplates/finorio-saas-landing/). Premium templates count toward the total. Every leaf project — including everything nested under templates/premium/<provider>/ — is one template. The headline count in templates/README.md and the templates table in the root README.md must equal the number of leaf projects on disk: When you add or move a premium template, update that headline count and both README…
vanzan01CLAUDE.mdvanzan01/claude-code-sub-agent-collective/CLAUDE.md
This is the Claude Code Sub-Agent Collective - an NPX-distributed framework that installs specialized AI agents, hooks, and behavioral systems for TDD-focused development workflows. The system enforces test-driven development through automated handoff validation and provides intelligent task routing through a hub-and-spoke architecture. Git Remote URL: https://github.com/vanzan01/claude-code-sub-agent-collective.git NEVER CHANGE THIS URL - Always use this exact repository URL for all git operations. For testing changes before publishing (see ai-docs/Simple-Local-Testing-Workflow.md): MANDATORY NAMING CONVENTION: All npm testing directories MUST follow the established pattern:
vanzan01CLAUDE.mdvanzan01/claude-code-sub-agent-collective/templates/CLAUDE.md
No summary in the file. Open it to read it.
ink1ingCLAUDE.mdink1ing/anti-api/CLAUDE.md
- src/main.ts - CLI entry point - src/server.ts - Hono HTTP server setup - src/services/antigravity/chat.ts - Core chat logic - src/proto/encoder.ts - Protobuf encoding with model selection - Antigravity requests use the provider API directly; the old local language-server port discovery path is intentionally not part of the runtime. Models are specified via model parameter in requests. See MODEL_ENUM in encoder.ts for supported values. Supports /v1/messages, /v1beta/messages, and /messages endpoints for maximum compatibility. Hosted providers are wired through src/services/routing/router.ts (dispatch),…
posit-devCLAUDE.mdposit-dev/skills/CLAUDE.md
A collection of Claude Skills published by Posit PBC. Skills are structured markdown files that teach Claude specialized workflows (e.g., Shiny app development, R package testing, GitHub PR workflows). There is no application code to build, compile, or deploy — the primary artifacts are Markdown files consumed directly by Claude's skill system. The only runnable utility is count-skill-tokens.py, which reports line and token counts for a skill: Warns when SKILL.md exceeds 5,000 tokens / 500 lines, or when the skill…
muthuishereCLAUDE.mdmuthuishere/mcp-server-bash-sdk/CLAUDE.md
A Model Context Protocol server SDK in pure Bash, implementing MCP revision 2026-07-28 over both standard transports. Three layers, split by file: - mcpservercore.sh — the protocol layer. JSON-RPC framing, method dispatch, per-request version negotiation, result envelopes, error codes. It defines functions only; sourcing it starts nothing. runmcpserver is the stdio entry point. - mcpserverhttp.sh — the Streamable HTTP binding. Parses HTTP, enforces the header↔body contract, maps errors onto status codes, then calls **the same process_request**. It contains no protocol…
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.
