How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
MicrosoftAGENTS.mdmicrosoft/Data-Science-For-Beginners/translations/pcm/AGENTS.md
Follow di existing code style and patterns - Add explanatory comments for complex logic - Update relevant documentation - Test changes across different lesson modules if e apply - Review di CONTRIBUTING.md file ## Additional
BlockAGENTS.mdblock/buzz/AGENTS.md
they are defects authors agree with on sight. Authors apply them **before writing code**, and reviewers check against them; each cites the PRs where reviewers litigated it. 1. **Every caught
BlockAGENTS.mdblock/buzz/desktop/src/features/agents/AGENTS.md
code is worse than no rule; a new pattern that isn't written down here will be broken by the next agent that never learns it existed. Reviewers: treat
Hugging FaceAGENTS.mdhuggingface/diffusers/.ai/AGENTS.md
post to GitHub directly** — no comments, reviews, or replies. Draft when asked and hand the text to the user. ## Code formatting - `make style` and `make fix-copies` should
can1357AGENTS.mdcan1357/oh-my-pi/AGENTS.md
streams, temp files) | | `crates/pi-natives` | Rust crate for performance-critical text/grep ops | **Catalog import convention**: code in this repo imports catalog _values_ (bundled models, model-thinking helpers, identity, descriptors, model manager/cache
alibabaAGENTS.mdalibaba/nacos/AGENTS.md
intended behavior differs from the existing spec, do not silently implement the code first: discuss the change with maintainers and update the affected specs as part of the design
iOfficeAIAGENTS.mdiOfficeAI/AionUi/AGENTS.md
path aliases: `@/*`, `@process/*`, `@renderer/*` - Prefer `type` over `interface` (per Oxlint config) - English for code comments; JSDoc for public functions ### Internationalization (i18n) New or changed user-facing text must use i18n
OpenAIAGENTS.mdopenai/openai-python/AGENTS.md
party GitHub Actions to reviewed full commit SHAs. Minimize job-level token permissions and never expose secrets or write-capable tokens to untrusted pull-request code. - Preserve separate build
tirth8205AGENTS.mdtirth8205/code-review-graph/AGENTS.md
must push - If push fails, resolve and retry until it succeeds ## MCP Tools: code-review-graph **This project has a knowledge graph. Start with the code-review-graph MCP tools
VercelAGENTS.mdvercel-labs/agent-skills/skills/vercel-optimize/AGENTS.md
projects that are not deployed on Vercel, greenfield projects with no traffic, or general code review. ## Requirements - Node.js 20+ - Vercel CLI with `vercel metrics`, `vercel usage`, `vercel contract`, and `vercel
ComposioHQAGENTS.mdComposioHQ/composio/AGENTS.md
other AI coding agents. ## Scope This is the Composio SDK v3 monorepo. Most product code lives in `ts/` and `python/`; docs live in `docs/`. Branch new work from `next
ComposioHQAGENTS.mdComposioHQ/composio/docs/AGENTS.md
docs-decisions` skill for docs content, changelogs, decisions, docs automation, or docs review. - Context references live under `docs/agent-guidance/context/`; use `docs/agent-guidance/context/twoslash.md` before editing typed MDX examples. - Agent workflow prompts live under
LangChainAGENTS.mdlangchain-ai/deepagents/AGENTS.md
time context, not required startup reading. - Treat source code and tests as authoritative. A brief's unknowns and review items are verification gaps, not automatic requirements. - Prefer the narrowest quiet
LangChainAGENTS.mdlangchain-ai/deepagents/examples/deploy-coding-agent/AGENTS.md
before proceeding ### Phase 2: Implement - Follow your plan step by step - Write clean, idiomatic code that matches existing patterns - Run tests after each significant change - If tests fail, debug
OpenAIAGENTS.mdopenai/openai-agents-python/AGENTS.md
Contributor Guide ## 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
kestra-ioAGENTS.mdkestra-io/kestra/AGENTS.md
what a paragraph would tell them is already in the code. Write a comment only when a reviewer would otherwise have to ask **why** the code exists: - a non-obvious
BloopAIAGENTS.mdBloopAI/vibe-kanban/AGENTS.md
utils`, `git` (Git operations), `api-types` (shared API types for local + remote), `review` (PR review tool), `deployment`, `local-deployment`, `remote`. - `packages/local-web/`: Local React + TypeScript app entrypoint (Vite, Tailwind). Shell source
gastownhallAGENTS.mdgastownhall/beads/AGENTS.md
flagged. ## Key Sections - **Issue Tracking** - How to use bd for work management - **Development Guidelines** - Code standards and testing - **Project Scope** - Read [engdocs/PROJECT_CHARTER.md](engdocs/PROJECT_CHARTER.md) before adding new feature surface area - **Visual
manaflow-aiAGENTS.mdmanaflow-ai/cmux/cmux-browser/AGENTS.md
derived files retain GPL-3.0-only provenance and the exact source revision. Helium-derived code is not available under Manaflow's commercial license. - Other bundled dependencies retain their own terms
manaflow-aiAGENTS.mdmanaflow-ai/cmux/web/AGENTS.md
request. Do not add a broad disable or raise the limit to accept new code. Lower the limit in a separate cleanup wave as the remaining debt is removed
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.
