AGENTS.md examples from real projects

CloudflareAGENTS.md

circl

cloudflare/circl/AGENTS.md

CIRCL is an experimental, externally visible Go cryptographic library. Consumers include Cloudflare's Go fork, downstream modules, and external researchers. Every change is security-sensitive. The README disclaims API stability — the…

NVIDIAAGENTS.md

Personal-AI-Router

NVIDIA/Personal-AI-Router/AGENTS.md

Guidance for agents and contributors working in the NVIDIA Personal AI Router (PAIR) repository. It holds an Electron desktop application and the Go services it runs on, built from one checkout. This file is the short orientation. CONTRIBUTING.md is the contribution policy and Developer Guide is the code tour; both are normative where this file is only a summary. .cursor/rules/ is not advisory. Those files record the boundaries the code is actually held to — process ownership, IPC shape, typing…

CloudflareAGENTS.md

foundations

cloudflare/foundations/AGENTS.md

- Telemetry scope stacks must use std::thread_local! through ScopeStack; do not replace them with thread_local::ThreadLocal, whose metadata can cause false sharing on these hot paths. - UserSpan::deferred() creates a stable request root that can be captured by telemetry contexts before activate() makes the sampling decision. - TelemetryContext::with_user_span() attaches an explicit UserSpan without changing ambient scope; the returned context and handle share the same underlying span. - A deferred root is an inactive span in a shared RwLock. Activation replaces it…

VercelAGENTS.md

openreview / vercel-react-best-practices

vercel-labs/openreview/.agents/skills/vercel-react-best-practices/AGENTS.md

Version 1.0.0 Vercel Engineering January 2026 Note: This document is mainly for agents and LLMs to follow when maintaining, generating, or refactoring React and Next.js codebases. Humans may also find…

JetBrainsAGENTS.md

MPS

JetBrains/MPS/AGENTS.md

This project is a mixed JetBrains MPS + plain Java/Kotlin codebase. Developers 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. Use IDEA tools for JVM work and the full MPS MCP toolset for model work. Use this file as the cross-environment entry point. For any MPS work, load the mps-mcp-workflow skill; it holds…

Model Context ProtocolAGENTS.md

php-sdk

modelcontextprotocol/php-sdk/AGENTS.md

mcp/sdk is the official PHP SDK for the Model Context Protocol — server and client, framework-agnostic, PHP ^8.1. It is a PHP Foundation × Symfony collaboration and follows Symfony's coding…

No other project uses it yet

LangChainAGENTS.md

deepagentsjs

langchain-ai/deepagentsjs/AGENTS.md

When Corridor's analyzePlan tool is available, create a plan and use the tool to analyze it before generating or modifying code. Apply the resulting security guidance before writing code.

No other project uses it yet

LangChainAGENTS.md

deepagentsjs / memory

langchain-ai/deepagentsjs/examples/memory/AGENTS.md

This is a sample AGENTS.md file demonstrating the memory feature. This project uses a middleware-based architecture where:

No other project uses it yet

NVIDIAAGENTS.md

KAI-Scheduler

NVIDIA/KAI-Scheduler/AGENTS.md

KAI Scheduler is a Kubernetes scheduler optimized for GPU resource allocation in AI/ML workloads, built on kube-batch with a modular plugin architecture. E2E tests run against a real Kubernetes using kind cluster and are located in test/e2e/suites/.

PydanticAGENTS.md

httpx2

pydantic/httpx2/AGENTS.md

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

No other project uses it yet

MicrosoftAGENTS.md

OpenAPI.NET

microsoft/OpenAPI.NET/agents.md

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

MongoDBAGENTS.md

mongo-php-library

mongodb/mongo-php-library/AGENTS.md

The official high-level MongoDB PHP library (mongodb/mongodb). It provides a full-featured, typed API — clients, databases, collections, GridFS, change streams, bulk writes, codecs — on top of the low-level mongodb extension. The extension is a separate repository (mongodb/mongo-php-driver). It owns the wire protocol, connection management, and BSON encoding. This library builds the server commands and implements the operation-level behaviour on top. A change to how a command is executed or how BSON is encoded belongs in the extension; a change…

No other project uses it yet

Google DeepMindAGENTS.md

mujoco_warp

google-deepmind/mujoco_warp/AGENTS.md

See README.md for development setup. @CONTRIBUTING.md - PR body should be plain, concise prose. Describe the problem, what the change does, and any non-obvious tradeoffs. Bullet points listing changes are fine, but avoid section headers, structured templates, and emojis. A good PR description reads like a short paragraph to a colleague, not a form. - PR and commit messages are rendered on GitHub, so don't hard-wrap them at 88 columns. Let each sentence flow on one line.

No other project uses it yet

MicrosoftAGENTS.md

hve-core / references

microsoft/hve-core/.github/skills/hve-core/vally-tests/references/agents.md

Nine conformance checks the vally-tests skill emits for .agent.md artifacts (including consolidated subagent structural template), with contract citations, stimulus shapes, and Vally grader recommendations

VercelAGENTS.md

dev3000

vercel-labs/dev3000/AGENTS.md

This file is the source of truth for agent guidance in this repo. CLAUDE.md is symlinked here.

No other project uses it yet

VercelAGENTS.md

dev3000 / www

vercel-labs/dev3000/www/AGENTS.md

This version has breaking changes — APIs, conventions, and file structure may all differ from your training data. Read the relevant guide in node_modules/next/dist/docs/ (resolved from this file's directory; in monorepos the next package may not be visible from the repo root) before writing any code. Heed deprecation notices. This block is written and re-added by next dev — verify at node_modules/next/dist/server/lib/generate-agent-files.js. Removing it from a diff only re-creates the uncommitted change; committing it with your work keeps the tree…

Model Context ProtocolAGENTS.md

swift-sdk

modelcontextprotocol/swift-sdk/AGENTS.md

Guidance for AI coding agents, and the humans driving them, working on the Swift SDK for the Model Context Protocol. Read this file before changing code. AI agents should: -…

No other project uses it yet

MicrosoftAGENTS.md

playwright-java

microsoft/playwright-java/AGENTS.md

The Java client is a port of the JavaScript client in ../playwright/packages/playwright-core/src/client/. When implementing or changing a method, read the corresponding JS file first and mirror its logic. Project checkouts (including the upstream playwright repo) live in the parent directory (../). Use the gh cli to interact with GitHub. The tests are in playwright/src/test/java/com/microsoft/playwright/. A test is a @PlaywrightTest method that takes the fixtures it needs as parameters, resolved by type: Page page, BrowserContext context, Browser browser, BrowserType browserType, Playwright…

No other project uses it yet

MicrosoftAGENTS.md

playwright-java / test

microsoft/playwright-java/playwright/src/test/AGENTS.md

These rules say how to write playwright/src/test/java/com/microsoft/playwright/tests/{page,library} from tests/page and tests/library in microsoft/playwright (the checkout in ../playwright; the commit the port is based on is in upstream_base.txt, together with the commands that list the upstream changes still to port). Use them for every new or changed upstream test. The port is line by line, with zero deviations. A reviewer reads the .spec.ts file and the .java file side by side. Each upstream statement maps to one Java statement, or to…

No other project uses it yet

OpenAIAGENTS.md

openai-java

openai/openai-java/AGENTS.md

Read CONTRIBUTING.md and SECURITY.md before changes. .castiron.stats.yml records generation metadata. Follow Generated-file ownership before editing SDK files. Apply Security expectations for credentials, diagnostics, dependency/build integrity, action pins, publishing permissions, security review, and private disclosure. Additional requirements: Large application/json bodies and streaming events are normal Responses, Chat Completions, and other API output. Do not add arbitrary fixed body/event/line limits as security or efficiency fixes. Any new rejection limit requires an explicit owner-approved API contract and review of supported payloads/transports. Preserve longstanding…