CLAUDE.md examples from real projects

How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Worked for most · soon

Known publishers Hiding risky files

BlockCLAUDE.md

artifact-swap

block/artifact-swap/CLAUDE.md

@ai-rules/.generated-ai-rules/ai-rules-generated-AGENTS.md

1710mo agoReviewMount

BlockCLAUDE.md

qrgo

block/qrgo/.agents/CLAUDE.md

.agents/AGENTS.md

176mo agoReviewMount

AWSCLAUDE.md

n8n-nodes-agentcore

aws/n8n-nodes-agentcore/CLAUDE.md

This project keeps all agent instructions in AGENTS.md. Always follow the rules in ./AGENTS.md, and when working in subdirectories, also check for a closer AGENTS.md in that directory tree.

163mo agoReviewMount

NVIDIACLAUDE.md

dl-lowlat-infer

NVIDIA/dl-lowlat-infer/CLAUDE.md

Do not duplicate AGENTS.md content here. For workflows, see README.md.

163mo agoReviewMount

PydanticCLAUDE.md

claude-code-logfire-plugin

pydantic/claude-code-logfire-plugin/CLAUDE.md

A Claude Code plugin that captures sessions and exports pydantic-ai compatible OTel traces to Pydantic Logfire, with local JSONL fallback. The plugin is a single Python 3 script (stdlib only, no external dependencies). The entire plugin is a single Python script (scripts/log-event.py) invoked by every hook event defined in hooks/hooks.json. The plugin manifest lives at .claude-plugin/plugin.json. Data flow: Claude Code hook fires -> stdin JSON piped to log-event.py -> appends JSONL locally -> if LOGFIRE_TOKEN set and event is SessionStart/Stop/SubagentStop/SessionEnd,…

169mo agoReviewMount

VercelCLAUDE.md

oss-gtm-feedback / www

vercel-labs/oss-gtm-feedback/apps/www/CLAUDE.md

This is a Next.js 16 application for GTM feedback collection and reporting. It uses: This project uses next-safe-action v8 for type-safe server actions with a sophisticated middleware pipeline. Located in src/lib/actions/clients/, we have three main action clients:

1610mo agoReviewMount

MicrosoftCLAUDE.md

cates / ecosystem

microsoft/cates/examples/ecosystem/CLAUDE.md

No summary in the file. Open it to read it.

157d agoReviewMount

MicrosoftCLAUDE.md

cates / ecosystem

microsoft/cates/fixtures/ecosystem/CLAUDE.md

No summary in the file. Open it to read it.

157d agoReviewMount

MicrosoftCLAUDE.md

ppteval / claude-workspace

microsoft/ppteval/claude-workspace/CLAUDE.md

You are running inside a single-task workspace for the PowerPoint Computer Use benchmark. Each invocation handles exactly one task. The harness has prepared this directory with everything you need. 1. Read TASK.md and OUTPUT_INSTRUCTIONS.md first. They define the goal and the exact path you must write your output to. Follow the output path literally — the grader looks for that path and nothing else. 2. Treat inputs/ as read-only. Copy the file before modifying it. 3. Write your result to…

155mo agoReviewMount

Microsoft AzureCLAUDE.md

sap-automation-qa

Azure/sap-automation-qa/CLAUDE.md

No summary in the file. Open it to read it.

142mo agoReviewMount

BlockCLAUDE.md

chalkline

block/chalkline/CLAUDE.md

Read AGENTS.md and follow it for every task in this repository.

1454d agoReviewMount

BlockCLAUDE.md

model-ledger

block/model-ledger/CLAUDE.md

Open-source model inventory and governance framework. Apache-2.0. This is an Apache-2.0 open-source project. All code must be generic and useful to any organization. Organization-specific connectors, auth, and backends go in separate packages. Examples in code, tests, and docstrings must be generic — use names like "Credit Scorecard", "feature_pipeline", "risk-team", "scoring_model". Never use organization-specific system names, team names, queue names, or person names in OSS code. The project is designed to be wrapped by downstream packages that add organization-specific connectors, auth,…

145mo agoReviewMount

GoogleCLAUDE.md

lightflow

google/lightflow/CLAUDE.md

When orchestrating multi-step workflows with side effects (e.g., database migrations, API batch modifications, code deployments, or multi-stage reports), use Lightflow to guarantee topological execution order, atomic state checkpoints, and human-in-the-loop approval gates. When working via the bash tool in Claude Code (python3 -m lightflow is equivalent to lightflow whenever the console script is not on PATH): - Scaffolding new workflows (create_lightflow): When asked to build a new Lightflow workflow, run the create_lightflow meta-workflow that ships in the Lightflow repository to…

1427d agoReviewMount

MicrosoftCLAUDE.md

agent-resources

microsoft/agent-resources/CLAUDE.md

When the user's request matches an available skill, ALWAYS invoke it using the Skill tool as your FIRST action. Do NOT answer directly, do NOT use other tools first. The skill has specialized workflows that produce better results than ad-hoc answers. Key routing rules: - Product ideas, "is this worth building", brainstorming → invoke office-hours - Bugs, errors, "why is this broken", 500 errors → invoke investigate - Ship, deploy, push, create PR → invoke ship - QA, test the…

145mo agoReviewMount

SentryCLAUDE.md

launchpad

getsentry/launchpad/CLAUDE.md

Launchpad is a Python-based microservice for analyzing iOS and Android app bundle sizes. It provides detailed analysis of app binaries, generates optimization insights, and offers size breakdowns with visualization. This…

1313mo agoReviewMount

MicrosoftCLAUDE.md

cat

microsoft/cat/CLAUDE.md

When the user's request matches an available skill, invoke it via the Skill tool. When in doubt, invoke the skill. Key routing rules: - Product ideas/brainstorming → invoke /office-hours - Strategy/scope → invoke /plan-ceo-review - Architecture → invoke /plan-eng-review - Design system/plan review → invoke /design-consultation or /plan-design-review - Full review pipeline → invoke /autoplan - Bugs/errors → invoke /investigate - QA/testing site behavior → invoke /qa or /qa-only - Code review/diff check → invoke /review - Visual polish →…

136mo agoReviewMount

LangChainCLAUDE.md

macaw-design-system

langchain-ai/macaw-design-system/CLAUDE.md

No summary in the file. Open it to read it.

1255d agoReviewMount

MicrosoftCLAUDE.md

AgentPlatformAdvisor

microsoft/AgentPlatformAdvisor/CLAUDE.md

When the user's request matches an available skill, invoke it via the Skill tool. When in doubt, invoke the skill. Key routing rules: - Product ideas/brainstorming -> invoke /office-hours - Strategy/scope -> invoke /plan-ceo-review - Architecture -> invoke /plan-eng-review - Design system/plan review -> invoke /design-consultation or /plan-design-review - Full review pipeline -> invoke /autoplan - Bugs/errors -> invoke /investigate - QA/testing site behavior -> invoke /qa or /qa-only - Code review/diff check -> invoke /review - Visual polish ->…

123mo agoReviewMount

NVIDIACLAUDE.md

structured-data-models-client

NVIDIA/structured-data-models-client/CLAUDE.md

See AGENTS.md.

1218d agoReviewMount

PydanticCLAUDE.md

connect4

pydantic/connect4/CLAUDE.md

Connect Four game demo showcasing PydanticAI and Pydantic Logfire capabilities. Features human vs AI and AI vs AI gameplay with multiple LLM model support. Live at https://logfire.run. IMPORTANT: ALWAYS USE…

1212mo agoReviewMount

CLAUDE.md vs AGENTS.md

About CLAUDE.md

What is CLAUDE.md?

A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.

Where does it go?

At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.

What should it contain?

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.md or AGENTS.md?

Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.