requirement-clarify SKILL.md skillA
requirement-clarify from mingdui/ai-product-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add mingdui/ai-product-skills/requirement-clarify@git:20261013.51f741bPins git:20261013.51f741b 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/requirement-clarify/SKILL.md- Versions recorded
- 4
- Latest version
- git:20261013.51f741b, committed Oct 13, 2026
- First version recorded
- Oct 13, 2026
- Size
- 8,282 bytes
- Repository stars
- 12
- License
- a license file GitHub does not identify
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、聊天记录或一句话描述，或需求来源方只讲了「要什么」没讲「边界在哪」的情况；不用于评审已成文的 PRD（那一步用 grill-with-prd）。.
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
# 需求澄清
把「一句话需求」采访成可以写 PRD 的输入。需求不完整时，AI 不会停下来问，它会按自己的理解补全——所以缺口必须在落笔前问出来。
## 使用边界
- **适用**：需求还停留在口头、聊天记录或一句话描述；需求来源方（老板、业务方、客户）讲清了「要什么」，但没讲边界在哪。
- **不适用**：需求已经完全明确，直接落文档 → 用 `prd-write`；已有充分上下文，不需要采访 → 用 `context-to-prd`。
### 与 `grill-with-prd` 的区别（重要，两者不重复）
| | 本 Skill | `grill-with-prd` |
|---|---|---|
| 时点 | PRD **成文之前** | PRD **成文之后** |
| 方式 | **建设式**采访——把缺的信息补齐 | **对抗式**压测——逐条质疑已有文档 |
| 产出 | `clarify.md`（喂给写 PRD 的那一步） | 直接写回 `prd.md` |
两者不能互相替代：没有 `clarify.md` 就打不出有质量的 PRD；没有 `grill-with-prd`，写出来的 PRD 边界仍然是虚的。
## 开始前
1. **确定功能名**——小写英文短横线（`pricing-page`），目标目录 `docs/prd/<feature-name>/`。
用户给的常常是中文描述而不是名字，那就由你译一个，但这三条不能省：
- **有歧义就列 2~3 个候选让用户选**（「会员中心」可能是 `user-center` 也可能是
`member-center`），**不要自己定**
- **把最终取的名字回给用户**——它会变成目录名，后面每一步都靠它衔接，静默决定等于埋雷
- 只确定这一次。后面 `prd-write`、`grill-with-prd` 等直接看目录就能推断，不用再传
完整命名规则（长度、去修饰词、写入前自检）见 [`../product-prd-extract/`](../product-prd-extract/)。
2. **先查代码库，再问人**。能从现有代码、已有 PRD、`docs/product/` 下的体验基线里查到答案的，不要拿去问用户——问已知的事会消耗掉对方的耐心。
3. 把用户已经说过的内容整理成一份「已知」清单，采访从「已知」之外的地方开始。
## 采访规则
- **一次只问一个问题**，等用户回答后再问下一个。不要一次抛出一串问题。
- 每个问题都给出**你推测的默认答案**，让用户确认或纠正。这比开放式提问快得多，也更容易问出真实边界。
- 用户答不上来时，记为 `[UNKNOWN]` 并标注需要谁来决定，**不要替他编**。
…Read the whole file at its exact version.
How to install
mdr add mingdui/ai-product-skills/requirement-clarify@sha256:d7e8702fc685fc53The line at the top of this page pins the label git:20261013.51f741b, 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.
[](https://markdownregistry.com/a/art_zys5whciywvcf6dh)
0 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 4 versions of this file, all committed on Oct 13, 2026.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261013.51f741b latest | 2026-10-13 | 51f741b | 8,282 B | A | view · diff |
| git:20261013.447718f | 2026-10-13 | 447718f | 7,650 B | A | view · diff |
| git:20261013.be4849d | 2026-10-13 | be4849d | 5,021 B | A | view · diff |
| git:20261013.60e8978 | 2026-10-13 | 60e8978 | 4,190 B | A | view |
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 (8282 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
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_zys5whciywvcf6dh GET https://markdownregistry.com/api/v1/resolve?ref=mingdui/ai-product-skills/requirement-clarify GET https://markdownregistry.com/api/v1/blob/d7e8702fc685fc53ce25d72120bfe19df706a79f82456e6f68137b6fb1cba0c4
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.