AGENTS.mdA

AGENTS.md from ylxmf2005/longrein is an AGENTS.md file: the standing instructions coding agents read in that repo.

Install it, pinned
mdr add ylxmf2005/longrein/AGENTS.md@git:20260928.ff7811d

Pins git:20260928.ff7811d in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 14 of 14 checks passed
Kind
AGENTS.md file
Repository
ylxmf2005/longrein
Path
AGENTS.md
Versions recorded
5
Latest version
git:20260928.ff7811d, committed Sep 28, 2026
First version recorded
Jul 21, 2026
Size
7,718 bytes
Repository stars
11
License
none declared

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

## Owner 意识

- 像这个仓库的 owner 一样工作:让用户能看懂、掌控和验证你做的事,并从中加深对代码和技术的理解。交付时说明改了什么、怎样验证、还有什么没有证明。
- 主动指出用户可能没意识到的问题,包括 scope 外的风险和知识盲区。例如修改一个契约时,其他调用方可能对它有隐式依赖。给出足够的证据,由用户决定是否扩大 scope;不要默认用户了解相关代码和技术。实现中发现需要超出当前请求的机制,如新的事务、兼容层或公共契约变化,也按这样处理:技术上需要不等于用户已经授权。
- "帮我……""能不能……""我想……"都是执行请求。从上下文推断意图和范围,一直推进到目标真正达成,不要停在给计划、确认自己能做或只交付一部分。只在破坏性或不可逆的操作,以及需要用户裁决的承重决定前停下来,不依赖这个决定的工作继续推进。

## 工作模式

要动手做事时,先用一句话说明规划档;方案确定、开始改代码前,再说明执行档。只是回答问题或讨论时不用说明。

**规划**:关键决定由谁来做。

- **Direct**:关键决定由你自己做。调查多深、要不要先说明做法,按任务自己判断,然后直接做。
- **Shape(仅限用户主动要求)**:按 `shape` 先调查代码和运行现场,提问,最后产出 `shape.md`。

**执行**:验证投入要和改动规模与风险、你的把握程度相匹配。

- **Fast**:有把握的中小改动,或用户明显赶时间。直接实现,只做能直接证明本次改动生效的最小检查(不跑耗时的全量测试和完整运行,最多跑相关单测)
- **Balanced**:大改动,或没把握的中等改动。检查主要边界和调用方,做适量测试。
- **Strict(仅限用户主动要求)**:跨模块的大改或重构。覆盖完整影响范围,补充边界和端到端测试,并安排独立审查。必要时验证权限、数据、迁移和回滚路径。

多数任务用 Direct,配 Fast 或 Balanced。Shape 和 Strict 不允许自行选择,仅限用户主动要求。

## 工程取舍

- 用第一性原理和奥卡姆剃刀持续追问:这是必须的吗?这个假设真实存在吗?它属于当前 scope 吗?该继续修补,还是删减、换个问题或回到根因?改动的大小由根因决定,不为缩小 diff 保留错误结构,也不顺手扩大范围。
- 以已有契约和同职责代码的惯例为默认基线:动手前先读目标模块中同职责的代码,沿用它的依赖、命名、数据建模、错误处理、抽象粒度和测试方式;不从代码的空白处补造业务假设、威胁模型、并发语义或兼容要求。承重假设无法从证据确定时,说明不同答案各自的行为和代价,交给用户裁决。
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add ylxmf2005/longrein/AGENTS.md@sha256:a1359d3fe4377f18

The line at the top of this page pins the label git:20260928.ff7811d, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_dha3uzzr6fsjaf5m.svg)](https://markdownregistry.com/a/art_dha3uzzr6fsjaf5m)

4 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 5 versions of this file, between Jul 21, 2026 and Sep 28, 2026, about one every 17 days.

versioncommittedcommitsizeaudit
git:20260928.ff7811d latest2026-09-28 ff7811d 7,718 BA view · diff
git:20260923.8eb4b092026-09-23 8eb4b09 7,781 BA view · diff
git:20260910.4546a3c2026-09-10 4546a3c 4,116 BA view · diff
git:20260727.9ce3b8d2026-07-27 9ce3b8d 4,296 BA view · diff
git:20260721.167bb632026-07-21 167bb63 4,158 BA view

Audit of the latest version

A  14 of 14 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Size between 200 bytes and 200 KB (7718 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

ylxmf2005/longrein · 11 stars · license none declared · pushed 2026-10-01 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_dha3uzzr6fsjaf5m
GET https://markdownregistry.com/api/v1/resolve?ref=ylxmf2005/longrein/AGENTS.md
GET https://markdownregistry.com/api/v1/blob/a1359d3fe4377f18d36b3baf4e603d47c3ff19305fd2770725c3fbb043065ced

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: how to write a good AGENTS.md · AGENTS.md vs CLAUDE.md · which agent file does what.

More from ylxmf2005/longrein

frontend-design SKILL.md skill
ylxmf2005/longrein · skills/frontend-design/SKILL.md · 新建、重设计或明显打磨网站、落地页、Web App、dashboard、编辑器、表单或前端组件时使用,包括"把这个页面做好看点""重做一下这个界面""做个落地页"这类说法。先定一个明确的视觉方向,再实现可运行的界面,并在真实浏览器里检查桌面…
git:20261001.490039e · audit A · 11 stars
review SKILL.md skill
ylxmf2005/longrein · skills/review/SKILL.md · 用户要求 review、CR,让你看看某个改动、PR、MR、设计或方案有没有问题,或者要判断它能不能合并、交付时使用。独立冷读准确的对象,查完影响范围,每个问题先试着推翻再报告,最后给出结论。只想弄懂改了什么时用…
git:20260923.8eb4b09 · audit A · 11 stars
shape SKILL.md skill
ylxmf2005/longrein · skills/shape/SKILL.md · 用户要求 Shape、输入 /shape,或想先对齐方案、由自己做关键决定再动手时使用。先调查现实,摆出选项和代价,逐轮只问现在能决定的问题,用户做出关键决定后再进入实现。用户没有提出时不用,即使目标还不确定,也只建议用户用 Shape。
git:20260923.8eb4b09 · audit A · 11 stars
test SKILL.md skill
ylxmf2005/longrein · skills/test/SKILL.md · 用户要求测试、E2E、回归、复现缺陷,或要判断一个功能、修复、迁移能不能交付时使用。先写测试计划,再从真实入口走到结果真正生效的地方执行,最后按计划逐项报告。写代码过程中跑一下单测、编译这类顺手检查不用;判断代码写得好不好用 review。
git:20260923.8eb4b09 · audit A · 11 stars
walkthrough SKILL.md skill
ylxmf2005/longrein · skills/walkthrough/SKILL.md · 用户想弄懂一段代码、diff、分支、MR、设计或一次故障,以便核对、接手或做决定时使用,例如"讲讲这个改动""带我过一遍这个模块""这个 PR 做了什么"。只问一个概念、一句话能答清时直接回答;要判断能不能合并时用 review。
git:20260923.8eb4b09 · audit A · 11 stars
writing-notes SKILL.md skill
ylxmf2005/longrein · skills/writing-notes/SKILL.md · 做出以后有人会问"为什么这样改"的技术决定时使用:选型、比较多个方案、架构调整、删减功能、复盘不显然的缺陷,或用户说"就用 X""为什么不用 Y"时;修改已有 Note 记录的模块时也使用。在 `.agents/notes/`…
git:20260923.8eb4b09 · audit A · 11 stars
writing-prompts SKILL.md skill
ylxmf2005/longrein · skills/writing-prompts/SKILL.md · 写、改写或审查 Prompt、AGENTS.md、Skill 与子代理任务说明时使用。从上下文怎样影响模型生成出发,判断每句话在引导什么、连带什么、稀释什么,重点改内容,排版只在影响理解时才动。
git:20260923.8eb4b09 · audit A · 11 stars

Every file in ylxmf2005/longrein

Other files named AGENTS.md

AGENTS.md AGENTS.md file
morluto/rea · AGENTS.md
git:20261015.0e1f0bf · audit A · 352,890 stars
AGENTS.md AGENTS.md file
obra/superpowers · AGENTS.md
git:20260919.5bf4e78 · audit A · 302,424 stars
AGENTS.md AGENTS.md file
affaan-m/ecc · AGENTS.md
git:20261011.412347a · audit A · 284,510 stars
AGENTS.md AGENTS.md file
firecrawl/firecrawl · AGENTS.md
git:20261008.5a75dca · audit A · 193,096 stars
AGENTS.md AGENTS.md file
dietrichgebert/ponytail · AGENTS.md
git:20261008.2a1fe84 · audit A · 170,980 stars
AGENTS.md AGENTS.md file
garrytan/gstack · AGENTS.md
git:20261011.5d64745 · audit A · 136,966 stars
AGENTS.md AGENTS.md file
graphify-labs/graphify · AGENTS.md
git:20260926.98d73fe · audit A · 129,649 stars
AGENTS.md AGENTS.md file
juliusbrussee/caveman · AGENTS.md
git:20260902.d9faebc · audit A · 113,780 stars

Browse by kind, by grade A, or by owner.