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.
SentryCLAUDE.mdgetsentry/sentry-mcp/packages/mcp-core/src/internal/agents/tools/data/CLAUDE.md
This directory contains JSON files for OpenTelemetry semantic convention namespaces used by the search-events tool's embedded AI agent. Each JSON file represents a namespace and follows this structure: Most files…
MetaCLAUDE.mdfacebookresearch/ProgramBench/CLAUDE.md
ProgramBench evaluates whether LM-based SWE-agents can reverse-engineer black-box software systems. The workflow: take an open-source CLI tool (mostly Rust/Go), compile it into a Docker image with source removed, then have an LM agent re-implement it from scratch by interacting only with the binary. Behavioral tests (also LM-generated) score the re-implementation. Some behavioral tests are unreliable and are excluded from scoring. Each excluded test is recorded under branches.<hash>.ignored_tests[] in a task's tests.json, with one or more reasons[].id explaining why. All ignored…
MicrosoftCLAUDE.mdmicrosoft/fabric-toolbox/tools/MicrosoftFabricMgmt/CLAUDE.md
This document provides comprehensive instructions for AI-assisted development of the MicrosoftFabricMgmt PowerShell module located in tools/MicrosoftFabricMgmt/. This module provides a robust interface for managing Microsoft Fabric resources via the Fabric API. Module Purpose: Enterprise-grade PowerShell module for Microsoft Fabric resource management with excellent error handling, comprehensive logging, and outstanding documentation. Build System: Sampler - Modern PowerShell module scaffolding and build framework Key Frameworks: - PSFramework: Configuration management and logging (Documentation) - Pester: Unit and integration testing - PlatyPS: Help documentation…
MicrosoftCLAUDE.mdmicrosoft/fabric-toolbox/tools/MicrosoftFabricMgmtMCPServer/CLAUDE.md
This directory contains a Python-based Model Context Protocol (MCP) server that wraps the MicrosoftFabricMgmt PowerShell module and exposes its capabilities to AI assistants (Claude, GitHub Copilot, etc.). DO NOT edit the PowerShell module source here. All PowerShell module development belongs in ../MicrosoftFabricMgmt/source/ and must follow the module's own CLAUDE.md. Key principle: Python is a thin protocol-translation layer only. All Fabric API logic lives in the PowerShell module. Python does not call Fabric REST APIs directly. Maintains a single long-lived pwsh…
MongoDBCLAUDE.mdmongodb/mongo-php-driver/.claude/CLAUDE.md
This is the low-level MongoDB PHP extension (mongodb), written in C. It wraps libmongoc and libbson to expose a minimal PHP API for core driver functionality. Higher-level APIs live in the separate MongoDB PHP Library. Issues are tracked in the PHPC JIRA project. Commit messages and branch names reference JIRA ticket IDs (e.g., PHPC-2215). make test compiles the extension and loads it automatically — no need to run make all or make install first. For failed tests, all intermediate files…
AnthropicCLAUDE.mdanthropics/buffa/CLAUDE.md
@CONTRIBUTING.md See DESIGN.md for the architectural overview. If a change to buffa-codegen (notably message.rs, impl_message.rs, view.rs, oneof.rs, enumeration.rs, imports.rs) affects generated output, you must regenerate checked-in code before committing, or CI (check-generated-code) will fail: Most codegen changes don't touch descriptor-specific paths, so gen-wkt-types and gen-option-types are usually sufficient. A quick check: git status after task lint — if buffa-types/src/generated/ or buffa-proto-options/src/generated/ shows as modified, you forgot to commit the regen. Before producing a commit, run both review agents in parallel…
SupabaseCLAUDE.mdsupabase/wrappers/CLAUDE.md
Wrappers is a development framework for PostgreSQL Foreign Data Wrappers (FDWs), written in Rust. It enables querying external data sources (APIs, databases, files) as if they were regular PostgreSQL tables. The project is maintained by Supabase. Documentation: https://fdw.dev/ | API Docs: https://docs.rs/supabase-wrappers The project uses a single root Cargo workspace (Cargo.toml) containing supabase-wrappers, supabase-wrappers-macros, wrappers, and every crate under wasm-wrappers/fdw/. All members share one Cargo.lock and one MSRV. default-members is scoped to the native crates so a bare cargo build/cargo…
StripeReads credentialsCLAUDE.mdstripe/link-cli/CLAUDE.md
Link CLI — lets agents get secure, one-time-use payment credentials from a Link wallet. pnpm + Turborepo monorepo:
StripeCLAUDE.mdstripe/pg-schema-diff/CLAUDE.md
A declarative schema migration tool for PostgreSQL that computes the difference between two database schemas and generates minimal, optimized SQL to migrate from one to the other with zero-downtime where possible. Problem Solved: Developers declare their desired database schema in DDL files, and pg-schema-diff automatically generates safe, optimized migration SQL that minimizes downtime and locks. Key Features: - Computes diffs between schemas (DDL files, databases, or directories) - Generates SQL using native Postgres online operations (concurrent index builds, online constraint…
LangChainReads credentialsCLAUDE.mdlangchain-ai/deep-agents-from-scratch/CLAUDE.md
This repository contains educational materials for building deep agents from scratch using LangGraph. It demonstrates progressive agent architectures through a series of Jupyter notebooks, starting with basic TODO list functionality and advancing to full agents with file systems and subagent spawning. State Management (state.py) - DeepAgentState: Extends LangGraph's AgentState with todos and files - Todo: TypedDict for task tracking with status (pending/in_progress/completed) - file_reducer: Merges file dictionaries in state updates Virtual File System (file_tools.py) - ls(): List files in virtual…
VercelCLAUDE.mdvercel/next-devtools-mcp/CLAUDE.md
This is an MCP (Model Context Protocol) server that acts as a thin connector between AI coding assistants and a running Next.js dev server. It discovers running Next.js 16+ dev…
MicrosoftCLAUDE.mdmicrosoft/CopilotStudioSamples/CLAUDE.md
Microsoft Copilot Studio samples repo with a Just the Docs (Jekyll) site for navigation and search. Live site: https://microsoft.github.io/CopilotStudioSamples/ Solutions follow the PnP format: Front matter: README should include: - What the solution does and which Copilot Studio features it demonstrates - Screenshots in assets/ (reference as ) - Installation steps (import zip via make.powerapps.com > Solutions > Import) - Any connection references or environment variables to configure - Known issues or limitations Update authoring/solutions/README.md Contents table after adding. Add…
LangChainCLAUDE.mdlangchain-ai/deep_research_from_scratch/CLAUDE.md
This repository builds a comprehensive deep research system from scratch using LangGraph, progressing through 5 tutorial notebooks that demonstrate different components and patterns. The notebooks in notebooks/ are the source of truth and should be the ONLY files modified. The source code in src/deep_research_from_scratch/ is automatically generated from the notebooks using %%writefile magic commands. Here's how it works: The system implements a three-phase deep research workflow: This approach ensures that the interactive tutorials remain the authoritative source while automatically maintaining…
MongoDBCLAUDE.mdmongodb/docs/CLAUDE.md
It holds only the rules that apply everywhere. Each area of the repo has its own CLAUDE.md with the rules for working there. This is the MongoDB Documentation Monorepo. It…
MongoDBCLAUDE.mdmongodb/docs/code-example-tests/CLAUDE.md
Grove is the infrastructure for creating tested, snippeted code examples for MongoDB documentation. Writers create example files with Bluehawk markup, test files that verify the examples work, and expected output…
MongoDBCLAUDE.mdmongodb/docs/content/CLAUDE.md
Ground rules for AI agents working on documentation in content/. These supplement the repo-root CLAUDE.md, whose critical instructions, git, and PR rules also apply. File paths below are relative to…
MongoDBCLAUDE.mdmongodb/docs/platform/CLAUDE.md
Ground rules for AI coding agents working in platform/. These supplement the repo-root CLAUDE.md, whose critical instructions, git, and PR rules also apply. For setup, environment variables, and staging details,…
Hugging FaceReads credentialsCLAUDE.mdhuggingface/Repo2RLEnv/CLAUDE.md
Auto-loaded by Claude Code for anyone working in this repo. It is committed and shared, so it holds only what is true for any clone: architecture, invariants, and the mistakes that are easy to make here. It is not a changelog and not a contributor guide. Those live elsewhere: If anything here conflicts with the code, trust the code and fix this file. Repo2RLEnv (repo2rlenv on PyPI) turns a repository into a verifiable RL training/eval dataset. End-to-end: synthesis → standardize…
ShopifyCLAUDE.mdShopify/cli/.claude/CLAUDE.md
See @../AGENTS.md for repository-specific agent instructions.
Model Context ProtocolCLAUDE.mdmodelcontextprotocol/experimental-ext-skills/CLAUDE.md
The line above is a Claude Code import: it expands AGENTS.md into context at session start. A symlink would also work, but it requires Administrator privileges or Developer Mode on Windows, so the import is used instead. https://code.claude.com/docs/en/memory#agents-md Claude-specific instructions, if any are ever needed, go below the import. -->
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.
