AGENTS.md examples from real projects

docling-projectAGENTS.md

docling

docling-project/docling/AGENTS.md

communicate with the user. Use short sentences, active voice, and consistent terms. Keep code identifiers, API names, and quoted source text exact. State the result and any limits clearly. ## Project

No other project uses it yet

openinterpreterAGENTS.md

open-interpreter

openinterpreter/open-interpreter/AGENTS.md

your new functionality. Refactor existing code as necessary to make this happen. Likewise, when reviewing code, do not hesitate to push back on PRs that would unnecessarily add code

scikit-learnAGENTS.md

scikit-learn

scikit-learn/scikit-learn/AGENTS.md

This pull request includes code written with the assistance of AI. > The code has **not yet been reviewed** by a human. This is a **mandatory requirement**, not optional. Include

No other project uses it yet

prometheusAGENTS.md

prometheus

prometheus/prometheus/AGENTS.md

supporting analysis (Google Doc, issue, etc.) for complex changes. --- ## Code Style - Follow [Go Code Review Comments](https://go.dev/wiki/CodeReviewComments) and the formatting/style section of [Go: Best Practices for Production Environments

No other project uses it yet

facebookAGENTS.md

docusaurus

facebook/docusaurus/AGENTS.md

check its presence to know which rules to apply. ### General rules - Humans review all the code: always write code that is easy to read and review, following the [`code-style

No other project uses it yet

warpdotdevAGENTS.md

Warp

warpdotdev/Warp/AGENTS.md

feedback; defer checks and tests until after implementation where possible. 2. Once the code and self-review are complete, run the relevant tests and fix the code until they pass

traefikAGENTS.md

traefik

traefik/traefik/AGENTS.md

answer *why*, not *what*.** Comments that restate what the code already says are noise: they go stale and waste review time. Only add a comment when it records

No other project uses it yet

scrapyAGENTS.md

scrapy

scrapy/scrapy/AGENTS.md

only on what it has as well. - Review your own diff. Remove whatever the change does not need: unused parameters, dead code, leftover debugging, comments that restate the code. ## Code

No other project uses it yet

usememosAGENTS.md

memos

usememos/memos/AGENTS.md

response redaction. Black-box service tests live in `server/api/test/`, unit tests next to the code. ## Change Routing | Change | Update | Verify | | --- | --- | --- | | Go service or router behavior | Handlers under `server/`, rules under

No other project uses it yet

gitAGENTS.md

git / po

git/git/po/AGENTS.md

Reviewing translation quality ## Background knowledge for localization workflows Essential background for the workflows below; understand these concepts before performing any housekeeping tasks in this document. ### Language code and notation

withastroAGENTS.md

astro

withastro/astro/AGENTS.md

narrate change history ("now", "previously", "no longer") and never address the reviewer ("this correctly handles..."). State how the code works, not how it came to be or why the change

No other project uses it yet

ultralyticsAGENTS.md

ultralytics

ultralytics/ultralytics/AGENTS.md

intentionally removes them with evidence. **Review gate:** for every addition, the reviewer decides whether deleting or changing existing code would have fixed the problem instead — if it would, that

No other project uses it yet

tw93AGENTS.md

Pake

tw93/Pake/AGENTS.md

same source of truth. Local-only overrides (`CLAUDE.local.md`, `AGENTS.override.md`, `.claude/settings.local.json`) stay ignored. ## Code Conventions - No Chinese comments in any source (Rust / TypeScript / any file). Comments and identifiers in English; follow

No other project uses it yet

penpotAGENTS.md

penpot

penpot/penpot/AGENTS.md

Writing rules, from Orwell, 1946. These govern prose: docs, PR text, messages. Never touch code or technical terms; swap in everyday words only where precision survives. 1. Never

No other project uses it yet

chen08209AGENTS.md

FlClash

chen08209/FlClash/AGENTS.md

expose the same skills to Claude Code. Current skills cover localization, UI work, UI performance, core/platform changes, and pre-commit structural quality review

No other project uses it yet

rcloneAGENTS.md

rclone

rclone/rclone/AGENTS.md

elegant, smallest change you can. Do not refactor or re-order code unless necessary as this makes review more challenging. Re-use existing test scaffolding, existing or library routines

No other project uses it yet

spring-projectsAGENTS.md

spring-framework

spring-projects/spring-framework/AGENTS.md

usage. ## Docs - Reference docs are AsciiDoc in `framework-docs/modules/ROOT`; code snippets for `include-code::` live in `framework-docs/src/main/{java,kotlin}`. ## Commits and pull requests - Pull requests target `main

No other project uses it yet

ultralyticsAGENTS.md

yolov5

ultralytics/yolov5/AGENTS.md

intentionally removes them with evidence. **Review gate:** for every addition, the reviewer decides whether deleting or changing existing code would have fixed the problem instead — if it would, that

No other project uses it yet

appwriteAGENTS.md

appwrite

appwrite/appwrite/AGENTS.md

autoloading. Avoid dependencies outside the `utopia-php` ecosystem. Never hardcode credentials — use env vars. Code changes may require a container restart; logs live on the relevant container

No other project uses it yet

pingchesuAGENTS.md

gsd-hermes

pingchesu/gsd-hermes/docs/AGENTS.md

agents** detailed in this section. Twelve additional shipped agents (pattern-mapper, debug-session-manager, code-reviewer, code-fixer, ai-researcher, domain-researcher, eval-planner, eval-auditor, framework-selector, intel-updater