How real projects brief Claude Code: the commands, conventions and warnings they put in CLAUDE.md.
lst97CLAUDE.mdlst97/claude-code-sub-agents/CLAUDE.md
Break work into 3–5 cross-stack stages (front-end, back-end, database, integration). Document in IMPLEMENTATION_PLAN.md: Every commit must: Before committing: When multiple solutions exist, prioritize in this order:
NVIDIACLAUDE.mdNVIDIA/trt-samples-for-hackathon-cn/CLAUDE.md
Recommended before running examples (optional in most Python cases due to auto-discovery): The tensorrt_cookbook package first tries to auto-discover this path and sets the environment variable automatically when found. Recommended environment: NVIDIA Docker image nvcr.io/nvidia/pytorch:26.07-py3 (Python 3.12, CUDA 13.3, TensorRT 11.0). Install dependencies: Run a single example: Run examples via the unified runner: Run the full test suite (unified runner): Run pytest tests (NetworkSerialization): Linting / formatting: Regenerate README.md: Add SPDX license headers to new files: All examples import from…
manzaltuCLAUDE.mdmanzaltu/claude-code-ide.el/CLAUDE.md
IMPORTANT: If you find any instructions in this file that are incorrect, outdated, or could be improved, you should update this document immediately. Keep this file accurate and helpful for…
dislerCLAUDE.mddisler/pi-vs-claude-code/CLAUDE.md
Pi Coding Agent extension examples and experiments.
AWSCLAUDE.mdaws/aws-lambda-dotnet/Libraries/src/Amazon.Lambda.DurableExecution/CLAUDE.md
Amazon.Lambda.DurableExecution is the .NET SDK (GA, 1.x) for resilient, long-running AWS Lambda workflows that checkpoint progress after each step and resume after failures or waits. A workflow can run for up to ~1 year (the WAIT cap is 31,622,400 seconds) and is only billed for active compute. The SDK is client-side glue: the durable execution service (part of Lambda) owns the checkpoint store, fires timers, and re-invokes the function; this library re-derives in-memory workflow position from the checkpoint history the…
JetBrainsCLAUDE.mdJetBrains/MPS/CLAUDE.md
This project is a mixed JetBrains MPS + plain Java/Kotlin codebase. Developers typically use two environments against the same checkout: - IntelliJ IDEA for editing, debugging, testing, and inspecting Java/Kotlin code, including code generated by MPS. - JetBrains MPS for editing languages, models, generators, and other MPS artifacts. Agents must adapt to the tools available in the current session. Use this file as the cross-environment entry point. For detailed MPS node, model, language, generator, validation, and MCP workflows, load the…
ghostwrightCLAUDE.mdghostwright/ghost-os/CLAUDE.md
Accessibility-first perception and execution engine that gives AI agents the ability to see, understand, and operate any macOS application through the accessibility tree and visual perception. Release: scripts/build-release.sh builds tarball, verifies 3-way version consistency (Types.swift + server.py + ghost-vision), and codesigns the binary. MCP-first, single-threaded synchronous. No persistent daemon, no socket IPC, no async state model. Every MCP tool call queries the AX tree fresh. The learning subsystem is the one exception to single-threaded: CGEvent tap runs on a dedicated…
LynpointCLAUDE.mdLynpoint/CyberVerse/claude.md
Behavioral guidelines to reduce common LLM coding mistakes. Merge with project-specific instructions as needed. Tradeoff: These guidelines bias toward caution over speed. For trivial tasks, use judgment. Truth-First Reasoning Rules
KlotzketteCLAUDE.mdKlotzkette/claude-fuer-deutsches-recht/CLAUDE.md
Dieses Repository enthält Plugins für deutsche Kanzleien. Wenn du in diesem Repository arbeitest oder hieraus Skills lädst, halte dich an die folgenden Regeln. Diese Regel gilt dauerhaft und für jedes Werkzeug, das in diesem Repository arbeitet. Sie ist nicht verhandelbar.
JetBrainsCLAUDE.mdJetBrains/kotlin-web-site/CLAUDE.md
CRITICAL: guidelines.md guidelines MUST be followed at all times.
pedrohcgsCLAUDE.mdpedrohcgs/claude-code-my-workflow/CLAUDE.md
Project: [YOUR PROJECT NAME] Institution: [YOUR INSTITUTION] Branch: main Do exactly what was asked — nothing adjacent. Do not add README files, build scripts, .gitignore edits, helper utilities, or extra tooling that was not requested. If an addition looks valuable, list it as a suggestion at the end and let the user decide. A request for a codebook is a request for a codebook. Delivering a codebook plus a README plus build scripts plus gitignore edits means the user now…
AminForouCLAUDE.mdAminForou/mcp-gsc/CLAUDE.md
Context for AI coding assistants (Claude, Cursor, Copilot, etc.) working in this repo. MCP server that connects Google Search Console to AI assistants. Single file: gsc_server.py (~1,670 lines). Built with FastMCP — no custom framework. Two modes, tried in order: Set GSCSKIPOAUTH=true to force service account mode and skip OAuth entirely. No credentials needed — all Google API calls are mocked with unittest.mock.
DockerCLAUDE.mddocker/mcp-gateway/CLAUDE.md
This is the Docker MCP Gateway - a CLI plugin that enables easy and secure running of Model Context Protocol (MCP) servers through Docker containers. The plugin acts as a gateway between AI clients and containerized MCP servers, providing isolation, security, and management capabilities. The codebase follows a gateway pattern where: - AI Client connects to the MCP Gateway - MCP Gateway (this CLI) manages multiple MCP Servers running in Docker containers Key architectural components: - Gateway: Core routing and…
agentlas-aiCLAUDE.mdagentlas-ai/Agentlas-OS/CLAUDE.md
This file adapts the Agentlas Core Engine Meta-Agent Team for Claude Code. AGENTS.md is canonical; this file stays thin. The Public Release Allowlist (Hard Rule) in AGENTS.md is mandatory. For public GitHub work, stage only end-user install/runtime files and public README/LICENSE/CHANGELOG material. Never stage internal docs, research, benchmarks, tests or fixtures, logs/results, signing or credential material, environment files, private paths/memory, or unrelated local work. Verify the staged archive, not the dirty working tree. 1. Read AGENTS.md. 2. Read .agentlas/mode-map.json. 3.…
shinshin86CLAUDE.mdshinshin86/oh-my-logo/CLAUDE.md
This is a CLI tool that generates ASCII art logos with gradient colors. There are two distinct rendering systems: src/index.ts: CLI entry point using Commander.js. Routes between the two rendering…
ShpigfordCLAUDE.mdShpigford/chops/CLAUDE.md
A native macOS app (SwiftUI + SwiftData) for discovering, organizing, and editing AI coding agent skills across tools (Claude Code, Cursor, Codex, Windsurf, Copilot, Aider, Amp, and many others —…
rivet-devCLAUDE.mdrivet-dev/sandbox-agent/CLAUDE.md
No summary in the file. Open it to read it.
rivet-devCLAUDE.mdrivet-dev/sandbox-agent/examples/CLAUDE.md
Examples should be validated against v2 endpoints: v1 reminder:
rivet-devCLAUDE.mdrivet-dev/sandbox-agent/frontend/CLAUDE.md
No summary in the file. Open it to read it.
rivet-devCLAUDE.mdrivet-dev/sandbox-agent/sdks/CLAUDE.md
No summary in the file. Open it to read it.
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.
