recruit-grill skillA
recruit-grill is agent-read markdown (skill) from viy1204/recruiting-copilot: 逐岗逼问式梳理真实岗位需求:用一次一问的访谈把"想招什么人"从模糊说法逼成可执行的初筛标准, 产出对外 JD、对内寻源笔记(目标公司/命脉技能/排除信号/搜索词)、CONTEXT.md 硬规则与术语表更新。 当用户说"新开一个岗位"、"梳理 XX 岗的要求"、"帮我写 JD",或寻源数据显示现有标准跑偏需要重梳理时使用。.
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# recruit-grill —— 逐岗梳理真实岗位需求 把用人需求从「我要一个厉害的 XX」逼问成**可执行的初筛标准**。一次只梳理一个岗位。 ## 访谈纪律(最重要) - **一次只问一个问题**。问完等回答,再问下一个。一次抛一串问题会把用户问懵。 - **每个问题都给推荐答案**:基于已知信息给出"我建议是 X,因为 Y",让用户确认或纠正,比开放题快得多。 - **能自己查到的不问**:用户已有旧 JD、公司介绍、已梳理过的其他岗位(CONTEXT.md),先读再问,只问文件里没有的。 - **追问模糊词**:用户说"要资深的"、"能力强的"、"最好懂 AI",必须逼问成可判断的标准 (几年算资深?看什么信号算能力强?懂 AI 是会用工具还是能落地到工作流?是硬要求还是加分项?)。 - **沉淀术语**:访谈中出现公司黑话、岗位简称、内部代号,随手写进 CONTEXT.md「术语表」,之后统一用这些词。 ## 问题清单 按 [`references/question-bank.md`](references/question-bank.md) 的顺序走(有依赖关系,别乱序)。 核心七组:岗位存在意义 → 硬门槛 → 硬底子 vs 表层 → 命脉技能与验证方式 → 排除信号 → 目标公司与搜索词 → 渠道预判。 ## 产出(每岗四件,全部写完才算梳理完) 1. **对外 JD** `01-jd/<role>.md`(可以直接发给候选人/挂平台的版本)。结构: ``` # 岗位名 > 背景一句话(为什么招) ## 一、招聘 spec(设计依据,内部对齐用,发布时可裁掉) | 维度 | 结论 | ← 定位/核心存在意义/汇报线/学历年限/地点/薪资带宽 ## 二、岗位详情 | 汇报对象 | 工作地点 | 薪资 | > 💡 关于这个岗位(人话版描述) ### 岗位职责 ### 任职要求(硬性) ### 加分项 > 📌 作品集/代码/案例要求(如适用) ``` ⚠️ 对外 JD 里**不写**商业敏感信息和寻源策略。 2. **对内笔记** `01-jd/_internal/<role>.md`——套 `_shared/templates/jd-internal.md`: 硬约束一句话锚定、命脉技能与验证方式、目标公司锚点、参考简历信号、排除信号、空的关键词迭代表。 写完提醒用户:**这份不外发**。 3. **CONTEXT.md 更新**: - 「初筛硬规则」:本次访谈确认的通用硬门槛(年龄线/学历线/地点坐班/错位处理)。已有内容冲突时向用户确认后更新,并在「已对齐决策」记一条带日期的变更。 …
Read the whole file at its exact version.
How to install
mdr add viy1204/recruiting-copilot/recruit-grill@git:20260713.c537f53mdr add viy1204/recruiting-copilot/recruit-grill@sha256:d871084a327814b9Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_h3tw2cn6qrvuztqf)
1 badge views in 30 days
Versions
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (3647 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
viy1204/recruiting-copilot · 77 stars · license MIT · pushed 2026-09-21 · branch master
API
GET https://markdownregistry.com/api/v1/artifacts/art_h3tw2cn6qrvuztqf GET https://markdownregistry.com/api/v1/resolve?ref=viy1204/recruiting-copilot/recruit-grill GET https://markdownregistry.com/api/v1/blob/d871084a327814b9b1400a8c427e956ee96c4d988e9ab3c91d6605e8ac0ef0ed
Your agent 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.