grill-with-prd SKILL.md skillA

grill-with-prd from mingdui/ai-product-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add mingdui/ai-product-skills/grill-with-prd@git:20261013.447718f

Pins git:20261013.447718f in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
mingdui/ai-product-skills
Path
skills/grill-with-prd/SKILL.md
Versions recorded
3
Latest version
git:20261013.447718f, committed Oct 13, 2026
First version recorded
Oct 13, 2026
Size
4,687 bytes
Repository stars
12
License
a license file GitHub does not identify

针对 PRD 文档的逐项审查会话,挑战需求的完整性、一致性和可实现性,并将确认的决策同步回 docs/prd/{feature-name}/prd.md。适用于用户想在开始实现前压测 PRD 的场景。.

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

<要做什么>

逐一追问 PRD 中的每个需求和决策,直到双方对功能边界、实现路径和验收标准达成共识。每个问题等用户回答后再继续,不一次性抛出所有问题。

如果某个问题可以通过探索代码库来回答,先探索代码库,不要把它抛给用户。

</要做什么>

<支撑信息>

## PRD 文档位置

PRD 文件路径:`docs/prd/{feature-name}/prd.md`

开始前先读取对应的 PRD 文件。如果用户没有指定 feature-name,询问后再继续。

## 审查会话中的行为

### 挑战模糊需求

遇到含糊的用户故事或需求描述时,要求精确化。

> "你说'用户可以管理订单'——管理包含哪些操作?创建、编辑、取消、删除都有吗?"

### 追问边界条件

对每个核心功能,主动构造边界场景来压测需求是否完整。

> "支付失败时,订单状态如何变化?用户能重试吗?重试次数有限制吗?"

### 发现需求冲突

当 PRD 中不同部分存在矛盾时,立刻指出。

> "用户故事第 3 条说用户可以删除草稿,但实现决策里说所有操作都需要审计日志——删除操作也要记录吗?"

### 对照代码库验证

当用户描述某个功能的工作方式时,检查代码库是否与之一致。发现矛盾时明确指出。

> "你说用户登录后会话有效期是 7 天,但代码里的 token 过期时间是 24 小时——以哪个为准?"

### 追问范围外的内容

PRD 的"范围外"部分如果过于简略,追问是否遗漏了重要的排除项。

> "范围外只写了'不支持批量操作',权限管理呢?这个功能需要区分角色吗?"

### 确认验收标准

对每个用户故事,确认是否有明确的验收标准,否则无法判断实现是否完成。

> "这条用户故事没有验收标准——成功的定义是什么?有性能要求吗?"

## 「指出」不等于「裁决」

这是本 Skill 最容易越界的地方。压测到一半往往要整体重写 PRD,**重写时顺手把发现的冲突一并"解决"掉,看起来是效率,实际是越权。**

分清两类:

| 类型 | 例子 | 你能做的 |
| --- | --- | --- |
| **澄清** | R1 写「显示月付价」、R2 写「默认年付」——只是措辞不一致 | 直接改成一致,**并在交付里说一句改了什么** |
| **裁决** | R1 同时要求「不产生横向滚动」和「卡片可横向滚动」——二选一是个设计决策 | **必须回到用户。** 你可以把选项和后果摆清楚,不能替他选 |

属于业务取舍、设计选择、对外口碑的冲突,一律归第二类。

**如果确实在重写中做了非澄清性的选择**(有时上下文里已经有答案,来不及再问),
那就在交付说明里**单独列出**,别混在「本次审查的主要变更」里一笔带过——
否则用户会以为他点过头。
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add mingdui/ai-product-skills/grill-with-prd@sha256:41b24381a2713ca8

The line at the top of this page pins the label git:20261013.447718f, 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_mvx7zbekhmbv7z6w.svg)](https://markdownregistry.com/a/art_mvx7zbekhmbv7z6w)

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

Versions

The registry has recorded 3 versions of this file, all committed on Oct 13, 2026.

versioncommittedcommitsizeaudit
git:20261013.447718f latest2026-10-13 447718f 4,687 BA view · diff
git:20261013.be4849d2026-10-13 be4849d 3,377 BA view · diff
git:20261013.60e89782026-10-13 60e8978 2,748 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (4687 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

mingdui/ai-product-skills · 12 stars · license a license file GitHub does not identify · pushed 2026-11-05 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_mvx7zbekhmbv7z6w
GET https://markdownregistry.com/api/v1/resolve?ref=mingdui/ai-product-skills/grill-with-prd
GET https://markdownregistry.com/api/v1/blob/41b24381a2713ca85f9e67fe07646e8b8c473f0050622a9d18dce1fb6df1668d

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: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from mingdui/ai-product-skills

CLAUDE.md CLAUDE.md file
mingdui/ai-product-skills · CLAUDE.md
git:20261013.3d17e09 · audit A · 12 stars
acceptance-checklist SKILL.md skill
mingdui/ai-product-skills · skills/acceptance-checklist/SKILL.md · 基于 PRD 和交付包生成结构化验收清单,产出 docs/prd/<feature-name>/acceptance.md。按核心流程、边界与异常、UI 还原度、数据埋点四个维度拆解,每条标 P0/P1/P2 优先级并注明依据出处(引用…
git:20261013.447718f · audit A · 12 stars
context-to-prd SKILL.md skill
mingdui/ai-product-skills · skills/context-to-prd/SKILL.md · 把已经聊透的需求上下文直接落成 PRD,不做采访。适用于你已经跟 AI 把问题、边界、方案讨论清楚,只差成文的场景;需求还停在口头、边界没问清时改用 requirement-clarify,要复用已有产品体验基线或做增量 PRD 时用…
git:20261013.be4849d · audit A · 12 stars
design-spec-prototyper SKILL.md skill
mingdui/ai-product-skills · skills/design-spec-prototyper/SKILL.md · 将 `design.md` / `DESIGN.md` 等有证据支撑的设计规格,**配合设计稿图片或截图**,转为高保真、无障碍且响应式的 UI 原型。规格提供品牌令牌、组件规则与字体(持续沉淀的产品事实),图片提供本次页面的实际构图,两者…
git:20261013.447718f · audit A · 12 stars
pm-frontend-prototype SKILL.md skill
mingdui/ai-product-skills · skills/pm-frontend-prototype/SKILL.md · 仅当用户明确要求在已有前端项目中生成、制作、构建或完善“原型”(例如高保真原型、可点击演示或评审 Demo)时使用此 Skill。默认基于现有源码复用组件、设计变量和技术栈,创建与正式业务入口隔离的原型页面或路由,并使用页面级模拟数据完成可…
git:20261013.4d20990 · audit A · 12 stars
prd-write SKILL.md skill
mingdui/ai-product-skills · skills/prd-write/SKILL.md · 编写或维护项目内的 PRD。适用于首版/重大方案的完整 PRD,也适用于已有产品基线后的功能迭代 PRD;先复用 docs/product 下已确认的体验规则,只记录本次新增、偏离、候选模式和待决问题。输出保存到 docs/prd…
git:20261013.447718f · audit A · 12 stars
product-experience-foundation SKILL.md skill
mingdui/ai-product-skills · skills/product-experience-foundation/SKILL.md · 为一个具体产品项目建立或维护可持续演进的体验基线,包括产品上下文、C 端或 Admin 的信息架构、视觉与交互规则、页面模式和组件晋升记录。适用于首版…
git:20261013.447718f · audit A · 12 stars
product-prd-extract SKILL.md skill
mingdui/ai-product-skills · skills/product-prd-extract/SKILL.md · 把 Word(.docx)PRD 或粘贴的 PRD 文本转成结构化 Markdown,落到 docs/prd/<feature-name>/prd.md,直接接进本仓库的链路。当用户提到「解析 PRD」「提取需求文档」「PRD…
git:20261013.51f741b · audit A · 12 stars
prototype-html SKILL.md skill
mingdui/ai-product-skills · skills/prototype-html/SKILL.md · 生成交互式HTML产品原型的专家技能。当用户需要创建产品原型、UI界面原型、 可交互的HTML演示页面、产品功能展示页面,或提到"原型"、"prototype"、 "产品界面"、"UI原型"、"交互原型"、"演示页面"等时使用。
git:20261013.60e8978 · audit B · 12 stars
prototype-to-product SKILL.md skill
mingdui/ai-product-skills · skills/prototype-to-product/SKILL.md · 将已确认的 HTML、Tailwind、Figma 导出或代码原型路由为当前仓库的正确工程交付:直接实现生产代码,或交接到项目采用的 OpenSpec、Superpowers…
git:20261105.993e405 · audit A · 12 stars
requirement-clarify SKILL.md skill
mingdui/ai-product-skills · skills/requirement-clarify/SKILL.md · 在写 PRD 之前,用采访的方式把一句模糊需求补成可以落笔的完整输入。逐个问清目标角色、触发场景、范围、明确不做的部分和已知约束,产出 docs/prd/<feature-name>/clarify.md。适用于需求还停留在口头、聊天记录或…
git:20261013.51f741b · audit A · 12 stars
web-design-extractor SKILL.md skill
mingdui/ai-product-skills · skills/web-design-extractor/SKILL.md · 从公开网站或已获授权的网站 URL、一个或多个网页截图中提取有证据支撑、可直接用于实现的 `design.md`。适用于需要为复刻、设计系统文档或 UI 实现简报逆向分析网站的视觉系统、布局、组件、响应式行为、内容层级或交互状态的场景。
git:20261013.447718f · audit A · 12 stars

Every file in mingdui/ai-product-skills

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