AGENTS.md examples from real projects

How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.

Best matches · from page 20Worked for most · soon

Besty0728AGENTS.md

Unity-Skills

Besty0728/Unity-Skills/AGENTS.md

accepted boundary). Do not report "add auth / tighten CORS / validate Origin" as a security finding. - Optional-package modules (ProBuilder, XR, Netcode, YooAsset, DOTween, PrimeTween, Behavior, HybridCLR, Addressables, QFramework) detect their

1.8k12d agoDiscuss

VercelAGENTS.md

coding-agent-template

vercel-labs/coding-agent-template/AGENTS.md

This document contains critical rules and guidelines for AI agents working on this codebase. ## Security Rules ### CRITICAL: No Dynamic Values in Logs **All log statements MUST use static strings only

1.8k8mo agoReads credentialsDiscuss

open-mercatoAGENTS.md

open-mercato

open-mercato/open-mercato/AGENTS.md

carries `review` unless it is already in another pipeline state. - Category (`bug`, `feature`, `refactor`, `security`, `dependencies`, `enterprise`, `documentation`) and meta (`needs-qa`, `skip-qa`, `qa-approved`, `qa-self-verified

1.8k4mo agoReads credentialsDiscuss

ginlix-aiAGENTS.md

LangAlpha

ginlix-ai/LangAlpha/AGENTS.md

sibling files and MCP credentials. Workspace tool selection is a routing convention, not a security boundary; use separate computers for isolation. ### Prompts, memory & memos - **Prompts**: Jinja2 templates in `src/ptc_agent/agent/prompts/templates/`, config

1.8k5d agoReads credentialsDiscuss

ginlix-aiAGENTS.md

LangAlpha / desktop

ginlix-ai/LangAlpha/desktop/AGENTS.md

NAME` takes the certificate's name **without** the `Developer ID Application:` prefix that `security find-identity` prints, which is the whole string a person naturally copies. electron-builder picks

1.8k5d agoReads credentialsDiscuss

xiufengsunAGENTS.md

TokenTracker

xiufengsun/TokenTracker/AGENTS.md

insforge-debug`: diagnosing failures (SDK/HTTP errors, RLS denials, auth and OAuth issues) and running security or performance audits. - `insforge-integrations`: wiring external auth providers (Clerk, Auth0, WorkOS, Better Auth

1.8k2mo agoReads credentialsDiscuss

JamailarAGENTS.md

Beav

Jamailar/Beav/AGENTS.md

clear description, the affected areas (e.g., `desktop/electron/`), and any UI screenshots for renderer changes. ## Security & Configuration Tips - API keys and model settings are user-configured in the app settings

1.8k6mo agoDiscuss

happier-devAGENTS.md

happier

happier-dev/happier/AGENTS.md

capacity limits are path-specific contracts; establish each from current product evidence, a released security or compatibility obligation, or explicit approval. Cross-device reachability, execution placement, state ownership, persistence, availability

1.8k9d agoDiscuss

Google CloudAGENTS.md

scion / docs-site

GoogleCloudPlatform/scion/docs-site/AGENTS.md

hooks, observability). 7. **Technical Reference** — *Mode-agnostic; autogenerated from `reference/`.* CLI, API, config schemas, security. 8. **Contributing** — *Autogenerated from `contributing/`.* Architecture, ADK, harness-dev, runtime-dev, design catalog, logging

1.7k3mo agoDiscuss

CloudflareAGENTS.md

circl

cloudflare/circl/AGENTS.md

Consumers include Cloudflare's Go fork, downstream modules, and external researchers. Every change is security-sensitive. The README disclaims API stability — the bar is **"correct and justified,"** not "frozen." **When

1.7k4mo agoDiscuss

MicrosoftAGENTS.md

edgeai-for-beginners

microsoft/edgeai-for-beginners/AGENTS.md

size reduction with 85% performance retention - Real-time conversation capabilities with local models ### Security and Privacy - All processing happens locally - no data sent to cloud - Suitable for privacy-sensitive applications

1.7k11mo agoReads credentialsDiscuss

MicrosoftAGENTS.md

edgeai-for-beginners / en

microsoft/edgeai-for-beginners/translations/en/AGENTS.md

while retaining 85% performance - Enables real-time conversation capabilities with local models ### Security and Privacy - All processing is done locally - no data is sent to the cloud - Suitable for privacy

1.7k11mo agoReads credentialsDiscuss

MicrosoftAGENTS.md

edgeai-for-beginners / my

microsoft/edgeai-for-beginners/translations/my/AGENTS.md

Repository အကြောင်းအရာနှင့် သင်ကြားမှုလမ်းကြောင်းများ - **Study Guide**: [STUDY_GUIDE.md](STUDY_GUIDE.md) - သင်ကြားမှုအရင်းအမြစ်များနှင့် အချိန်ဇယား - **Support**: [SUPPORT.md](SUPPORT.md) - အကူအညီရယူရန် နည်းလမ်းများ - **Security**: [SECURITY.md](SECURITY.md) - လုံခြုံရေးပြဿနာများကို အကြောင်းကြားခြင်း ### Community Support - **GitHub Issues**: [အမှားများကို အကြောင်းကြားရန် သို့မဟုတ် အင်္ဂါရပ်များကို တောင်းဆိုရန်](https://github.com

1.7k11mo agoReads credentialsDiscuss

MicrosoftAGENTS.md

edgeai-for-beginners / pcm

microsoft/edgeai-for-beginners/translations/pcm/AGENTS.md

still keep 85% performance - E fit do real-time conversation wit local models ### Security and Privacy - All processing dey happen for local - no data dey go cloud - E good

1.7k11mo agoReads credentialsDiscuss

relaticleAGENTS.md

relaticle

relaticle/relaticle/AGENTS.md

=== .ai/relaticle/architecture rules === # Architecture Relaticle is a modular monolith: the CRM core lives in `app

1.7k7d agoReads credentialsDiscuss

qufei1993AGENTS.md

skills-hub

qufei1993/skills-hub/AGENTS.md

validation; verify the PR and return its link. Do not merge without authorization. ## Security Red Lines - Token、密码和私钥只能存入系统安全凭据存储，禁止进入数据库、配置文件、日志、URL 或同步仓库。 - 只有用户主动操作或明确开启的后台功能才能读取凭据；页面加载、Tab 切换、状态展示和普通启动不得读取。 - 开发版必须使用独立的凭据命名空间；授权

1.7k3d agoDiscuss

kwaroranAGENTS.md

Risuai

kwaroran/Risuai/AGENTS.md

1.7k41d agoDiscuss

VercelAGENTS.md

openreview / vercel-react-best-practices

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

invoked directly. Next.js documentation explicitly states: "Treat Server Actions with the same security considerations as public-facing API endpoints, and verify if the user is allowed to perform a mutation

1.7k7mo agoDiscuss

bubbuildAGENTS.md

bub

bubbuild/bub/AGENTS.md

pytest`, docs build if relevant) - docs updates when CLI behavior, commands, or architecture changed ## Security & Configuration Tips - Use `.env` for local secrets; never commit credentials. - Bub runtime settings are driven

1.7k14mo agoReads credentialsDiscuss

legelingAGENTS.md

PromptHub

legeling/PromptHub/AGENTS.md

confirmed user decisions. Ask only when an unresolved choice materially affects scope, user data, security, ownership, or compatibility. An explicit user decision updates the authoritative requirement immediately; record any implementation

1.7k4mo agoDiscuss

CLAUDE.md vs AGENTS.md

About AGENTS.md

What is AGENTS.md?

An open format for instructions to coding agents, read by Codex, Cursor and others. Think of it as a README written for agents.

Where does it go?

At the repository root, with more specific files in subdirectories. Agents read the one closest to the file they're editing.

What should it contain?

Setup and test commands, code style, and the rules a new contributor would need to know.

Does Claude Code read it?

Claude Code reads CLAUDE.md. A one-line CLAUDE.md that points at AGENTS.md covers both.