How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
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.
AWSCLAUDE.mdawslabs/Log-Analyzer-with-MCP/CLAUDE.md
Log-Analyzer-with-MCP is an MCP (Model Context Protocol) server that provides AI assistants access to AWS CloudWatch Logs for searching, analysis, and cross-service correlation. The --stateless flag enables stateless HTTP mode for Amazon Bedrock AgentCore integration.
Google CloudCLAUDE.mdGoogleCloudPlatform/gke-mcp/CLAUDE.md
@./AGENTS.md
AnthropicCLAUDE.mdanthropics/oncall-kit/CLAUDE.md
If STACK.md does not exist at the repo root, this is a fresh install and nothing has been configured. Before doing anything else in this session, greet the user with exactly this shape (adapt the words, keep the content): This repo is the on-call kit — playbooks, comms, and a self-improving incident log for running Claude-assisted on-call in your team's Slack channel. Nothing is active yet: no watching, no schedules, no paging. One thing to know up front: this agent…
ElasticCLAUDE.mdelastic/ml-cpp/CLAUDE.md
This file provides domain knowledge for AI coding assistants working on the ml-cpp repository. It consolidates the detailed rules in .cursor/rules/ into a single reference. For full details, see the…
GitHubCLAUDE.mdgithub/gh-aw-firewall/CLAUDE.md
awf (Agentic Workflow Firewall, package @github/awf) is a CLI that wraps any command in a sandboxed Docker network. It provides L7 (HTTP/HTTPS) egress control using Squid proxy, restricting network access to a whitelist of approved domains while giving the agent access to the host workspace and selected system paths via chroot and selective bind mounts. The system is orchestrated by src/cli.ts and managed by src/docker-manager.ts. Squid, the primary agent, and the general API proxy are the baseline services; private-repository enclaves…
MicrosoftCLAUDE.mdmicrosoft/simplechat/CLAUDE.md
SimpleChat is a Flask web application using Azure Cosmos DB, Azure AI Search, and Azure OpenAI. It supports personal, group, and public workspaces for document management and AI-powered chat. Every…
NVIDIACLAUDE.mdNVIDIA/simready-foundation/CLAUDE.md
Repo-local skills live under the agent-agnostic skill tree: Compatibility links: When updating a SimReady skill, edit the skills source of truth, including bundled references/, assets/, evals/, and assets/openai.yaml metadata. Helper scripts used by skills live under assets/scripts/ in this repository. Use the same repo-level agent guidance as Codex in AGENTS.md.
AWSCLAUDE.mdawslabs/nx-plugin-for-aws/CLAUDE.md
Read the following before making changes: Install dependencies: Run tests and update snapshots: Before raising a PR, validate changed generators in an example workspace using locally compiled versions:
JetBrainsCLAUDE.mdJetBrains/databao-agent/CLAUDE.md
Always include the ticket ID in commits: Move the associated YouTrack ticket to Review state using update_issue. When you finish implementing changes: Wait for user confirmation before proceeding. If not…
Microsoft AzureCLAUDE.mdAzure/sap-automation/CLAUDE.md
Canonical project instructions and cross-agent SDAF skills. @.github/copilot-instructions.md
AWSCLAUDE.mdawslabs/visual-asset-management-system/backend/CLAUDE.md
Auto-loaded when Claude Code works within backend/. Authoritative guide for Lambda handler development, Pydantic model creation, authorization enforcement, DynamoDB access, and validators. Testing details: backend/tests/CLAUDE.md. Copy-paste skeletons for a new handler / model / test file: backend/HANDLER_TEMPLATES.md. Update this tree when adding new handler domains, model files, or test directories (root CLAUDE.md Rule 11).
AWSCLAUDE.mdawslabs/visual-asset-management-system/backendPipelines/CLAUDE.md
This is the Claude Code steering document for the backendPipelines/ tree. It covers the pipeline architecture, S3 output-path conventions, assetId threading, and the end-to-end checklist for adding a new processing pipeline. For project-wide context, see the root CLAUDE.md. VAMS supports four pipeline execution types (PIPELINE_EXECUTION_TYPES in backend/backend/models/pipelines.py): Lambda (sync or async invocation of a Lambda function), SQS (async message to an SQS queue), EventBridge (async event to an EventBridge bus), and DeadlineCloud (an AWS Deadline Cloud job). SQS and EventBridge…
AWSCLAUDE.mdawslabs/visual-asset-management-system/backend/tests/CLAUDE.md
Auto-loaded when Claude Code works within backend/tests/. Covers pytest configuration, mock module hierarchy, per-handler conftest patterns, and event-shape conventions for backend unit tests. For handler patterns, see backend/CLAUDE.md. The root backend/conftest.py performs critical setup before every test: tests/mocks/ contains stand-in modules that are loaded in place of the real implementations. Layout mirrors the handler package tree: When adding a new dependency that a handler imports at module load time, add a matching mock module here if the real one requires…
AWSCLAUDE.mdawslabs/visual-asset-management-system/CLAUDE.md
This is the root-level Claude Code steering document for VAMS. It is auto-loaded in every session and provides the project-wide context that all agents need. For component-specific details, see the CLAUDE.md files in each subdirectory. VAMS is an AWS-native Visual Asset Management System for managing, visualizing, and processing 3D assets, point clouds, CAD files, and other visual content. It deploys as a CloudFormation/CDK stack with: VAMS version: see infra/config/config.ts and tools/VamsCLI/vamscli/version.py. Python 3.12 (Lambda), 3.13+ (dev). Node 22.x (Lambda). npm…
AWSCLAUDE.mdawslabs/visual-asset-management-system/documentation/CLAUDE.md
Steering document for Claude Code when working within the documentation/ directory. Auto-loaded when the working context is within documentation/. VAMS documentation is built with Docusaurus (React-based static site generator) and lives in documentation/docusaurus-site/. The source Markdown files are in documentation/docusaurus-site/docs/. Follow AWS documentation standards:
AWSReads credentialsCLAUDE.mdawslabs/visual-asset-management-system/infra/CLAUDE.md
This is the Claude Code steering document for the infra/ directory. It is auto-loaded when Claude Code operates within the VAMS CDK infrastructure-as-code. Maintenance note: Update this tree when adding new nested stacks, lambda builders, constructs, or pipeline types. See root CLAUDE.md Rule 11.
AWSCLAUDE.mdawslabs/visual-asset-management-system/infra/lib/nestedStacks/pipelines/CLAUDE.md
Auto-loaded when Claude Code operates within infra/lib/nestedStacks/pipelines/. Covers pipeline stack layout, required Lambda package layout in backendPipelines/, VPC builder wiring, sub-process and log registration wiring, and S3 output path conventions. See infra/CLAUDE.md for cross-stack patterns (lambda builder, service helper, security helpers). Each pipeline follows a consistent structure: CRITICAL — Pipeline Lambda Directory Structure: Every pipeline's lambda/ directory in backendPipelines/ MUST include: Without __init__.py and customLogging/logger.py, Lambda will fail at import time with No module named 'customLogging'. Copy these files from…
AWSReads credentialsCLAUDE.mdawslabs/visual-asset-management-system/tools/VamsCLI/CLAUDE.md
Steering document for Claude Code when working in the VamsCLI Python CLI tool. Auto-loaded when the working context is within tools/VamsCLI/. VamsCLI is a Python command-line interface built with the Click framework for interacting with the Visual Asset Management System (VAMS) deployed on AWS. It provides authentication, configuration management, multi-profile support, and full CRUD operations against the VAMS API Gateway. Maintenance note: Update this tree when adding new command groups, utility modules, or auth providers. See root CLAUDE.md Rule 11.
AWSReads credentialsCLAUDE.mdawslabs/visual-asset-management-system/tools/VamsMCP/CLAUDE.md
Steering document for Claude Code / Kiro when working in the VAMS MCP server. Auto-loaded when the working context is within tools/VamsMCP/. VamsMCP is a Model Context Protocol server that exposes the VAMS REST API as agent-callable tools. It is built with the mcp SDK (FastMCP) and reuses the vamscli package for API access and authentication. - Entry point: vams_mcp/server.py (main() -> mcp.run(), stdio transport) - Stores no credentials: authenticates via the user's existing vamscli profile (vamscli setup + vamscli…
AWSReads credentialsCLAUDE.mdawslabs/visual-asset-management-system/web/CLAUDE.md
This is the frontend-specific Claude Code steering document for the VAMS (Visual Asset Management System) web application. It is auto-loaded when working within the web/ directory. VAMS frontend is a React 18.3 + TypeScript ^5.0.0 single-page application built with Vite. All source files in src/ are .ts/.tsx (only remaining .js files are Jest mocks in src/__mocks__/). Primary UI library: AWS Cloudscape Design System (@cloudscape-design/components ^3.0.196). Package manager: npm. NEVER use yarn in this project. Key design decisions: Maintenance note: Update…
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.
