VercelCLAUDE.mdnext-docs-agentic-rag
vercel-labs/next-docs-agentic-rag/CLAUDE.md
No summary in the file. Open it to read it.
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
VercelCLAUDE.md
ElasticCLAUDE.mddocs-utils
elastic/docs-utils/CLAUDE.md
This repository builds elastic-docs-utils, a cross-harness installer for Elastic documentation skills and MCP configuration. Do not overwrite an existing MCP configuration or skill directory unless this project previously recorded ownership. Update checks must remain cache-only in host lifecycle hooks; network refreshes are explicit user commands.
Google CloudCLAUDE.mdmcp-python-sdk-grpc-poc
GoogleCloudPlatform/mcp-python-sdk-grpc-poc/CLAUDE.md
This document contains critical information about working with this codebase. Follow these guidelines precisely. Where <name> is the name of the user. - NEVER ever mention a co-authored-by or similar aspects. In particular, never mention the tool used to create the commit message or PR. - Create a detailed message of what changed. Focus on the high level description of the problem it tries to solve, and how it is solved. Don't go into the specifics of the code unless…
MicrosoftPipes a download into a shellCLAUDE.mdaibast-agents-library
microsoft/aibast-agents-library/CLAUDE.md
AIBAST Agents Library is the stable Microsoft downstream for the RAPP (Rapid Agent Prototype Pattern) stack. It combines the local Brainstem server, an industry agent-template catalog, the RAPP production guide, Azure deployment assets, and a Power Platform solution. Philosophy: "engine, not experience" — this is infrastructure, not a consumer product. See CONSTITUTION.md for scope rules and governance. The durable beta product direction is documented in beta/GOLDEN_PATH.md: - RAPP accelerates learning AI and teaching AI immediately after learning it. - The…
MicrosoftReads credentialsCLAUDE.mdaibast-agents-library / project
microsoft/aibast-agents-library/community_rapp/project/CLAUDE.md
Read CONSTITUTION.md first. It defines what belongs in this repo and what doesn't. CommunityRAPP is a base memory agent platform built on Azure Functions with Azure OpenAI integration. Think of it as a "brainstem" — a minimal, extensible runtime that gives any AI agent persistent memory across conversations. The platform provides: - Two built-in agents: ContextMemory (read) and ManageMemory (write) - Dual-layer memory: shared (all users) + user-specific (per GUID) - Five HTTP endpoints, including grail-compatible and legacy conversation routes…
MicrosoftReads credentialsCLAUDE.mdaibast-agents-library / rapp_ai
microsoft/aibast-agents-library/rapp_ai/CLAUDE.md
Read CONSTITUTION.md first. It defines what belongs in this repo and what doesn't. CommunityRAPP is a base memory agent platform built on Azure Functions with Azure OpenAI integration. Think of it as a "brainstem" — a minimal, extensible runtime that gives any AI agent persistent memory across conversations. The platform provides: - Two built-in agents: ContextMemory (read) and ManageMemory (write) - Dual-layer memory: shared (all users) + user-specific (per GUID) - Three HTTP endpoints: health check, main conversation, and Copilot…
MicrosoftReads credentialsCLAUDE.mdaibast-agents-library / rapp_brainstem
microsoft/aibast-agents-library/rapp_brainstem/CLAUDE.md
RAPP Brainstem is a local-first AI agent server. It's a Flask app that proxies conversation through the GitHub Copilot API with automatic tool-calling via user-defined agents. Single dependency: a GitHub account with Copilot access. Runs on port 7071. Philosophy: "engine, not experience" — infrastructure only, no opinionated UI or workflows (see CONSTITUTION.md). No build step, linter, or type checker is configured. Entry point: brainstem.py — a single-file Flask server (~2,000 lines) that handles auth, chat, agent orchestration, and the web…
MicrosoftCLAUDE.mdomnichannel-amsclient
microsoft/omnichannel-amsclient/CLAUDE.md
Two-mode client selected by framedMode config flag: FramedClient loads a hidden iframe ({baseUrl}/{version}/iframe.html) containing IframeCommunicator. All AMS API calls execute inside the iframe. Host ↔ iframe communication via postMessage with request/response pattern and UUID-based request correlation. FramedlessClient calls API.ts functions directly via fetch. API layer (API.ts): Five functions — skypeTokenAuth, createObject, uploadDocument, getViewStatus, getView. Auth via skype_token header. AMS endpoint from chatToken.amsEndpoint || chatToken.regionGTMS?.ams.
AWSCLAUDE.mdagentic-developer-platform
awslabs/agentic-developer-platform/CLAUDE.md
For deployment, follow the canonical agent deployment guide. It owns account confirmation, phase ordering, private journals, recovery, verification and when to request user input. Use the quickstart for source and published-release commands. The references below supplement those instructions. Read docs/adp-platform-deployment/deployment-manifest.md for the complete mapping of every resource to its AWS service, module, and validation command. Use it to validate each phase after completion. Three modules on a shared AWS platform: Shared infrastructure: platform/infra/ (VPC, EKS, ECR, IAM). Follow deploy-with-agent.md for…
ElasticCLAUDE.mdobservability-migration-platform
elastic/observability-migration-platform/CLAUDE.md
Claude-specific guidance. For automation/agent rules see AGENTS.md. For public docs see docs/README.md. This repo (the Observability Migration Platform, CLI obs-migrate) is the canonical source for the migration engine — grafana-migrate,…
Model Context ProtocolReads credentialsCLAUDE.mdactions
modelcontextprotocol/actions/CLAUDE.md
This repo holds composite GitHub Actions reused across MCP-family repos. Before adding a new action, confirm the logic is (or will be) needed in more than one caller repo — single-repo logic should stay in that repo's workflow. Each action lives in its own directory. If a feature has multiple related entry points (e.g. deploy + cleanup), group them under a common parent: or for multi-action groups: Caller references: - modelcontextprotocol/actions/<action-name>@main - modelcontextprotocol/actions/<group-name>/<sub-action>@main Required fields Every action.yml must include: -…
PostHogCLAUDE.mdposthog-vscode
PostHog/posthog-vscode/CLAUDE.md
VS Code extension that brings PostHog into the editor: feature flags, experiments, event analytics, and session replay — all inline in your code. See ROADMAP.md for v1 scope, features, and what's been dropped.
ShopifyCLAUDE.mdbundler-ignore-dependency
Shopify/bundler-ignore-dependency/CLAUDE.md
This document provides guidance for LLMs and developers working on this project. bundler-ignore-dependency is a Bundler plugin that allows developers to ignore version constraints on Ruby, RubyGems, or gem dependencies. This is useful in monorepo scenarios where you want to test gems against specific versions without strict version constraints. The GitHub Actions workflow (.github/workflows/ci.yml) automatically: - Fetches the latest 2 major Bundler versions dynamically - Tests against Ruby 3.3, 3.4, and 4.0 - Runs RuboCop linting - Runs all 86…
StripeReads credentialsCLAUDE.mdstripe-commercetools-checkout-app
stripe/stripe-commercetools-checkout-app/CLAUDE.md
Two initialization paths: - Session flow (_Setup): fetches /config-element/payment + /operations/config + /customer/session using x-session-id header - Express no-session flow (_SetupExpress): fetches POST /express-config (CORS-only, no session needed) Supported builders: - createDropinBuilder('embedded') → Stripe Payment Element - createExpressBuilder('dropin') → Stripe Express Checkout Element - createComponentBuilder(type) → individual components (currently empty)
AWSCLAUDE.mdindustrial-shopfloor-connect
awslabs/industrial-shopfloor-connect/CLAUDE.md
NEVER run any git command that changes repository state. Git is the user's business, not yours. Specifically, do NOT: - create, switch, rename, or delete branches (git branch, git checkout -b, git switch) - stage, commit, or amend (git add, git commit) - merge, rebase, cherry-pick, or reset (git merge, git rebase, git cherry-pick, git reset) - push, pull, or fetch (git push, git pull, git fetch) - stash, tag, or otherwise mutate refs or the working tree via git
BlockCLAUDE.mdbuilder-syndicate / alpha-claude
block/builder-syndicate/specs/alpha-claude/claude.md
This project uses a premium purple-themed design system matching modern SaaS aesthetics. The design prioritizes depth, polish, and smooth interactions over flat minimalism. Spacing System (use --sp-X for new code): Backwards Compatibility: Old --space-X variables are aliased to --sp-X for legacy mockups. Color Gradients (key feature of premium design): Shadow System (creates depth): Border Radius: Required Structure: Key Features: - Gradient brand icon with shadow + glow - Glassmorphism: backdrop-filter: blur(20px) saturate(180%) - User points displayed in nav (gold color)…
SentryReads credentialsCLAUDE.mdskrooge
getsentry/skrooge/CLAUDE.md
Skrooge is a CLI tool for estimating Kubernetes scaling costs on GCP. Given a deployment's replica count, CPU/memory changes, and instance type, it calculates how many additional GCP instances are needed and their cost across different pricing tiers. skrooge tui does the same interactively, for several workloads at once. Repository: getsentry/skrooge Language: Python 3.11+ License: Apache 2.0 The project is managed with uv and built with the uv_build backend. This creates .venv with the package (editable) and the dev group.…
MicrosoftReads credentialsCLAUDE.mdvulcan
microsoft/vulcan/CLAUDE.md
Guidance for Claude Code (claude.ai/code) working in this repository. VULCAN generates multi-turn tool-calling training data from API specifications. It simulates the software a tool catalog describes entirely in memory, builds a dependency graph over the tools, drives conversations against that live environment, judges them, and emits training records. Public release of a codebase developed internally against a single hosted model endpoint, accompanying VULCAN: Where Agents Learn by Living in Simulated Tool Environments (DATA-FM @ ICLR 2026, <https://openreview.net/forum?id=Ht4HB4pIQq>). The paper's experiments…
PostHogReads credentialsCLAUDE.mdllm-analytics-apps
PostHog/llm-analytics-apps/CLAUDE.md
Internal tooling for the PostHog LLM Analytics team. Provider examples have been moved to the SDK repos (posthog-python and posthog-js) as examples/example-ai-* directories. What remains here: - Demo data generator (scripts/generate_demo_data.py) — generates realistic multi-turn conversations using PostHog AI SDKs - Trace generator (trace-generator/) — creates mock LLM trace data without real API calls - Test scripts (scripts/test_*.py) — integration tests for specific SDK features - Example runner (run-examples.sh) — discovers and runs examples from sibling SDK repos - Screenshot…
StripeCLAUDE.mdclaude-code-patches
stripe/claude-code-patches/CLAUDE.md
This is a project that is solely a GitHub workflow, a shell script, and GitHub releases. The goal is to:
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.