How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
Google CloudCLAUDE.mdGoogleCloudPlatform/professional-services/tools/agent-eval/CLAUDE.md
them 1. **AI generation** (recommended): `agent-eval init` → "Generate with AI". Gemini analyzes agent code, picks managed metrics, generates 2-4 binary `custom_llm_judge` metrics tailored to the agent
KhazPCLAUDE.mdKhazP/vibe-coding-prompt-template/templates/CLAUDE.md
AGENTS.md](./AGENTS.md).** Read it first. This file holds only what is specific to Claude Code, so the two never drift apart — duplicating commands or architecture across both is how they
jakubkrehelCLAUDE.mdjakubkrehel/make-interfaces-feel-better/CLAUDE.md
Claude Code (claude.ai/code) when working with code in this repository. ## What this repository is A standalone agent skill for reviewing and improving interface polish, published for installation
coderCLAUDE.mdcoder/claudecode.nvim/CLAUDE.md
split. - `keep_terminal_focus` (boolean, default: `false`) - When enabled, keeps focus in the Claude Code terminal when a diff opens instead of moving focus to the diff buffer. This allows
rlaopeCLAUDE.mdrlaope/oh-my-hermes/CLAUDE.md
move makes the next tracker run re-evaluate a range already reviewed. The receipt schema, the reason codes, and the baseline enrolment for rows that predate the contract
ax-llmCLAUDE.mdax-llm/ax/docs/CLAUDE.md
types** | `field:image/audio/file/url` | `photo:image "Profile picture"` | | **Code** | `field:code "description"` | `script:code "Python code"` | ### Advanced Field Usage ```typescript // Optional fields with string syntax const optionalSig = s(` userInput:string -> response
VRSENCLAUDE.mdVRSEN/OpenSwarm/CLAUDE.md
servers to an agent - `.cursor/commands/mcp-code-exec.md` - how to convert an MCP server into the Code Execution Pattern (progressive tool disclosure, 98% token reduction) - `.cursor/commands/write-instructions.md` - how to write effective instructions
kaitrannttCLAUDE.mdkaitranntt/ccs/CLAUDE.md
this file. ## Scope CCS is a TypeScript/Bun CLI and dashboard for managing Claude Code, Codex, Factory Droid, CLIProxy, and compatible provider profiles. ## Non-Negotiables - Default branch is `dev`. Feature/fix branches
tradermontyCLAUDE.mdtradermonty/claude-trading-skills/CLAUDE.md
export ALPACA_PAPER="true" # or "false" for live trading # Configure Alpaca MCP Server in Claude Code settings # See portfolio-manager/references/alpaca-mcp-setup.md for detailed setup guide ``` #### API Pricing and Access **Financial Modeling
compozyCLAUDE.mdcompozy/compozy/CLAUDE.md
procedure resolves a task-specific need; start with the owning skill. - Go conventions: `eng-code-guidelines`; Go tests: `eng-test-conventions`. Use `eng-consolidate-test-suites` when placement is unclear
fuergaosi233CLAUDE.mdfuergaosi233/claude-code-proxy/CLAUDE.md
successful software development tasks. ## Task Management For complex or multi-step tasks, Claude Code will use: * **TodoWrite**: To create a structured task list, breaking down the work into manageable steps
addyosmaniCLAUDE.mdaddyosmani/web-quality-skills/CLAUDE.md
CLAUDE.md This file provides guidance to Claude Code when working with this repository. ## Project summary Web Quality Skills is a measurement-first collection of Agent Skills for Google Lighthouse, Chrome
jeremylongshoreCLAUDE.mdjeremylongshore/tons-of-skills-marketplace/CLAUDE.md
gitleaks` + `skill-conform`** (GitHub Actions app; `strict:false`, `enforce_admins:false`, 0 required approving reviews, code-owner reviews not required). CODEOWNERS routes ownership and optional review
jgravelleCLAUDE.mdjgravelle/jcodemunch-mcp/CLAUDE.md
findings-row`), and a typed number is a review finding. ⚠⚠ **A competitor's README figure is never a measurement and competitor code runs only in the sandbox**; a release title
BytePioneer-AICLAUDE.mdBytePioneer-AI/codex-host/CLAUDE.md
implicit cross-module coupling. ## Code Size & Structure - Keep handwritten production modules focused on one primary responsibility. - Treat 500 lines as a design-review signal, not a hard limit. When
Hugging FaceCLAUDE.mdhuggingface/OpenEnv/CLAUDE.md
claude/skills/` and run inline: **Review & Validation Skills:** | Skill | Trigger | Definition | |-------|---------|------------| | [`alignment-review`](.claude/skills/alignment-review/SKILL.md) | "review this code" | Two-tier review (bugs + alignment flags) | | [`pre-submit-pr`](.claude/skills/pre-submit-pr/SKILL.md) | "ready for PR?" | Comprehensive
chrisryugjCLAUDE.mdchrisryugj/korean-law-mcp/CLAUDE.md
v4.14.1: `law_system`·`amendment_track`·`ordinance_compare`·`procedure_detail` 추가, 공용 `withChainDeadline` 헬퍼). `document_review` 는 기반 법령 프리픽스가 없고 갈래 수가 고정이라 개별 fetch 한도(전체 45초)만 건다
Natively-AI-assistantCLAUDE.mdNatively-AI-assistant/natively-cluely-ai-assistant/CLAUDE.md
change, and architectural decision must preserve correct behavior on both operating systems. --- ## MCP Tools: code-review-graph **IMPORTANT: This project has a knowledge graph. ALWAYS use the `code-review-graph
Nano-CollectiveCLAUDE.mdNano-Collective/nanocoder/CLAUDE.md
ToolManager.initializeCustomTools()` registers them into the same registry as built-ins and MCP tools — downstream code (`/tools`, subagents, mode filtering) sees them through the unified registry. Mode policy for custom tools
antonbabenkoCLAUDE.mdantonbabenko/terraform-skill/CLAUDE.md
code block / table adds a fact not in surrounding prose - [ ] Subsection under 400 tokens - [ ] Anchors referenced from SKILL.md remain stable - [ ] For substantive new sections, get a second-opinion review from
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.
