How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
OpenAIAGENTS.mdopenai/openai-agents-js/AGENTS.md
repository root. ## Table of Contents 1. [Policies & Mandatory Rules](#policies--mandatory-rules) 2. [Code Review Rules](#code-review-rules) 3. [Project Structure Guide](#project-structure-guide) 4. [Operation Guide](#operation
raullenchaiAGENTS.mdraullenchai/Rapid-MLX/AGENTS.md
plus tests proportional to risk. 5. Review the diff for regressions, generated files, secrets, and unrelated edits. 6. Record durable knowledge in code, tests, or docs; do not leave
nickjvandykeAGENTS.mdnickjvandyke/opencode.nvim/AGENTS.md
dispatched as `OpencodeEvent: ` User autocmds. OpenCode v2 events are shaped `{ id, type, data }`. - **Edit review**: opens diff in new tab via `:diffpatch`, keymaps `da`/`dr` to accept/reject
yvgudeAGENTS.mdyvgude/lean-ctx/AGENTS.md
given file or shared worktree at a time. Parallel agents investigate, test, or review in isolation unless explicit file ownership or separate worktrees are assigned. - **Completion:** the lead integrates
AtmosphereAGENTS.mdAtmosphere/atmosphere/AGENTS.md
expect(true).toBe(true)` is forbidden. ### Correctness Review Checklist Before committing, verify these for every changed file: 1. **Ownership** — Does this code close/release/shutdown only resources it created? 2. **Terminal paths
tutti-osAGENTS.mdtutti-os/tutti/AGENTS.md
recovered)? > > - Yes -> it must live in `packages/agent/host` (tuttid and tsh write only > delegate/adapter code). > - No (transport, DTO, query, presentation, product policy) -> adapter. > - Unsure -> answer in the PR description: "Does
tutti-osAGENTS.mdtutti-os/tutti/packages/AGENTS.md
tutti` or `__tsh`, and product bridge methods out of `browser/*` - do not move code into `packages/` until there is a real multi-consumer boundary - treat a public npm package
MicrosoftAGENTS.mdmicrosoft/mcp/AGENTS.md
dashes) - Prefer file-scoped changes over project-wide modifications when possible - Always review your own code for consistency, maintainability, and testability - Always ask for clarifications if the request is ambiguous
xintaofeiAGENTS.mdxintaofei/codeg/AGENTS.md
when working with code in this repository. ## 项目概述 Codeg（Code Generation）是一个多智能体编码工作台，它将多个智能体（Claude Code、Codex CLI、OpenCode、Gemini CLI、OpenClaw、Cline 等）统一到一个工作区中，支持会话聚合和多智能体协作，支持桌面安装，服务器/Docker 部署。 ## 技术栈
geekjourneyxAGENTS.mdgeekjourneyx/md2wechat-skill/AGENTS.md
shipped paths clearly. 3. If a feature is not backed by current code or release assets, do not describe it as shipped. 4. Prefer short, operational instructions over marketing copy
ObservalAGENTS.mdObserval/Observal/AGENTS.md
policy.** No deprecation wrappers. When code moves, callers update in the same PR. Dead code is deleted immediately. - **Tests mock externals.** No Docker needed to run the test suite
ObservalAGENTS.mdObserval/Observal/web/AGENTS.md
session traces** with conversation replay, span tree, and token counts - **Admin operations** such as review queue, user management, insights, and audit logs - **Agent building** with component assembly and live YAML
stickerdanielAGENTS.mdstickerdaniel/linkedin-mcp-server/AGENTS.md
synthetic prompt that would reproduce the PR diff if given to a fresh Claude Code session. Don't copy the user's first message — distill the conversation into a single
NVIDIAAGENTS.mdNVIDIA/TransformerEngine/AGENTS.md
bash qa/L0_pytorch_lint/test.sh TE_PATH="$PWD" bash qa/L0_jax_lint/test.sh ``` Both lint scripts also inspect common code. Use `CPP_ONLY=1` or `PYTHON_ONLY=1` to select one half during iteration. ## Documentation Update
graykodeAGENTS.mdgraykode/abtop/AGENTS.md
Policy English is mandatory for all project-facing work and communication. - Write all source code, comments, tests, fixtures, documentation, examples, configuration text, scripts, and user-facing strings in English
Ar9avAGENTS.mdAr9av/obsidian-wiki/AGENTS.md
along with the pending English diff — then translate and backfill those changes into `README_TW.md`. Reviewers assess translation quality. ## Configuration Resolve config using the Config Resolution Protocol in `llm-wiki/SKILL.md
OpenAIAGENTS.mdopenai/openai-go/AGENTS.md
described in `SECURITY.md`; never disclose them in public issues, pull requests, or discussions. ## Code Review Rules - Use `$openai-go-pr-review` for an exhaustive review of a pull request, committed
letta-aiAGENTS.mdletta-ai/letta-code/AGENTS.md
runs on `letta_agent_v3.py`. `letta_agent_v2.py` is summarization retry wrapper. Don't mix up naming in code review. - **Token counting differs by provider.** Anthropic: `input_tokens` excludes cached (total = input + cache_creation
AWSAGENTS.mdawslabs/agentcore-samples/01-features/02-host-your-agent/01-runtime/04-coding-agents/03-code-agents-competition-e2e/coding_agents/codex/AGENTS.md
user's message. 2. Use the MCP tools to read the issue, fix the code, and submit a PR. 3. Execute the requested action — do NOT just describe what
AWSAGENTS.mdawslabs/agentcore-samples/01-features/02-host-your-agent/01-runtime/04-coding-agents/03-code-agents-competition-e2e/coding_agents/open-code/AGENTS.md
human review. - NEVER close an issue. Leave issues open for the reviewer. - Always add the label `agent:open-code` to every issue and PR you touch. - Always add labels
An open format for instructions to coding agents, read by Codex, Cursor and others. Think of it as a README written for agents.
At the repository root, with more specific files in subdirectories. Agents read the one closest to the file they're editing.
Setup and test commands, code style, and the rules a new contributor would need to know.
Claude Code reads CLAUDE.md. A one-line CLAUDE.md that points at AGENTS.md covers both.
