How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
MicrosoftAGENTS.mdmicrosoft/Security-101/translations/tl/AGENTS.md
references - Mga Larawan: I-imbak sa `/images/` directory, gumamit ng deskriptibong alt text - Mga Code Blocks: Gumamit ng triple backticks na may language identifier kung naaangkop ### Mga Alituntunin sa Nilalaman
olimorrisAGENTS.mdolimorris/codecompanion.nvim/AGENTS.md
search. - When the user asks to fix tests, fix the tests - not the source code - unless explicitly asked otherwise. - If you're working with directories or files, utilise the functions
Devin-AXISAGENTS.mdDevin-AXIS/iPolloWork/AGENTS.md
video, say so explicitly and explain why, and include the exact commands/steps for the reviewer to reproduce. ## Validate Every Experience Almost everything we change has an effect on the outside
deanpetersAGENTS.mddeanpeters/Product-Manager-Skills/AGENTS.md
calls, and be documented in the skill. - Review skills and scripts for safety before sharing or running. - Do not delete existing Claude Code marketplace files such as `.claude-plugin/marketplace.json
steipeteAGENTS.mdsteipete/agent-scripts/AGENTS.MD
context` for direct API context preflight and `$autoreview` for isolated reviews; preserve its reviewer isolation. - Claude Code parallel/background work (Codex workers, monitors, long jobs): each = own harness-tracked task
MicrosoftAGENTS.mdmicrosoft/vscode-docs/docs/agents/concepts/agents.md
whether to take another action. 1. **Review:** When the task is complete, you review the response, code changes, and validation results. The repeated reasoning, action, and validation steps form
Human-Agent-SocietyAGENTS.mdHuman-Agent-Society/reef/AGENTS.md
accommodate new violations. ### Code structure and readability Apply these rules when writing or reviewing code. The [Code Review Style Guide](https://github.com/zhaochenyang20/sglang-diffusion-routing/issues/32) is a reference; the rules here take
op7418AGENTS.mdop7418/CodePilot/AGENTS.md
DISABLED=1`）**。**不自动跑 `test:smoke` / `test:e2e`**——那两个需要 dev server，按风险手动触发。 ## Codex review 规则 - 给 Claude Code 的执行文案必须共享判断过程：先写用户问题和争议，再写取舍理由，最后才写执行清单。不能只把聊天结论压成命令——ClaudeCode 重启或上下文变短后会重复旧误判。 - P1/P2 finding 不能只用聊天确认关闭，必须有修复、测试证据或 tech
SentryAGENTS.mdgetsentry/XcodeBuildMCP/AGENTS.md
errors from outdated dependencies; upgrade the dependency instead - Always ask before removing functionality or code that appears to be intentional - Do not add fallback behavior by default. If required context
MicrosoftAGENTS.mdmicrosoft/agent-governance-toolkit/agent-governance-opencode/AGENTS.md
OpenCode's contract: - `deny` → `throw new Error(...)` from the hook - `review` → annotate `output.args.__agt_review_reason` - `allow` → return undefined - exposes `agt_policy_status` and `agt_policy_check_text` as both
MicrosoftAGENTS.mdmicrosoft/agent-governance-toolkit/AGENTS.md
surfaces meant for direct external consumption | | Python runtime and product code | `*/` | Runtime code, applications, policy engines, trust, SRE, compliance, and other Python product surfaces that have not moved into
MicrosoftAGENTS.mdmicrosoft/agent-governance-toolkit/docs/AGENTS.md
discoverability at least once. - Security-sensitive changes should review `docs/security/` and `docs/security/threat-model.md`. ## Validation - Check that links, file paths, and fenced code blocks are valid. - Keep headings and page titles stable
MicrosoftAGENTS.mdmicrosoft/aspire/AGENTS.md
Multi-platform support (Windows, Linux, macOS, containers) ## General * Make only high confidence suggestions when reviewing code changes. * Always use the latest version C#, currently C# 13 features. * Never change global.json
rullerzhou-afkAGENTS.mdrullerzhou-afk/clawd-on-desk/AGENTS.md
# AGENTS.md This file is the entry point for coding agents working in this repository
uditgoenkaAGENTS.mduditgoenka/autoresearch/AGENTS.md
tests) 3 c3d4e5f 88.3 +1.2 keep add error handling tests ``` --- ## Agent-Specific Notes ### Claude Code - Commands are invoked as `/autoresearch` and `/autoresearch: ` - Interactive setup uses `AskUserQuestion` when context is missing
jacob-bdAGENTS.mdjacob-bd/gemini-notebook-mcp-cli/AGENTS.md
core/ # Low-level API client (no business logic) │ ├── client.py # Internal batchexecute API calls │ ├── constants.py # Code-name mappings (CodeMapper class) │ └── auth.py # AuthManager for profile-based token caching └── utils/ ├── config.py # Configuration
Q00AGENTS.mdQ00/ouroboros/AGENTS.md
Analysis**: analysis-agent, ontology-analyst, codebase-explorer, research-agent, breadth-keeper **Execution**: code-executor, seed-closer, consensus-reviewer, semantic-evaluator **Specialist**: advocate, judge
UfoMiaoAGENTS.mdUfoMiao/zcf/AGENTS.md
pnpm test` / `pnpm test:run` — execute Vitest suites; add `:coverage` for V8 coverage reports. ## Coding Style & Naming Conventions Stick to TypeScript ESM modules with 2-space indentation and single quotes
huangruitengAGENTS.mdhuangruiteng/loopx/AGENTS.md
contradictions require rule repair plus negative or mutation coverage. - Characterize before moving code. For status, quota, review-packet, scheduler, monitor, and handoff behavior, add or extend parity fixtures first, then
ElasticAGENTS.mdelastic/go-elasticsearch/AGENTS.md
AGENTS.md This file provides guidance to AI coding agents working with code in this repository. CLAUDE.md is a symlink to this file — they are the same file, do not update
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.
