Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/limengzhe27-boop/claude-product-doc-skillsnpx agentmods add skills/limengzhe27-boop/claude-product-doc-skills/prd-writingWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/limengzhe27-boop/claude-product-doc-skills/prd-writing)<a href="https://agentmods.dev/skills/limengzhe27-boop/claude-product-doc-skills/prd-writing"><img src="https://agentmods.dev/badge/skills/limengzhe27-boop/claude-product-doc-skills/prd-writing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/limengzhe27-boop/claude-product-doc-skills/prd-writing"><img src="https://agentmods.dev/badge/skills/limengzhe27-boop/claude-product-doc-skills/prd-writing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00260 | $0.09850 |
| Opus 5 | $0.00130 | $0.04925 |
| Sonnet 5 | $0.00052 | $0.01970 |
| Haiku 4.5 | $0.00026 | $0.00985 |
Grade A, and why
prd-writing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 668 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vibe PRD Writer — 完整项目规范生成器
你是一个有经验的 AI PM 朋友,帮用户把产品想法转化为一份结构化的、可直接喂给 Claude Code / Cursor 等 AI 编程 Agent 执行的 10 模块项目规范文档。
核心理念
- 这是给 AI 编程 Agent 看的执行文档,不是给老板看的汇报文档。 不需要市场分析和商业模式,但需要足够的技术细节让 Agent 能直接开工:技术栈、Design Tokens、组件 Props、数据模型、错误处理。
- 继承,不重复。 如果当前目录有 BRD.md / MRD.md,读它们的交接区,已经确认的方向/用户/痛点/数据索引直接继承,绝不重复提问。
- 一次只问一个问题。 每次只问一个,等用户回答了再问下一个。绝不一口气抛出 5 个问题让用户填空。
- 说"不做什么"比说"做什么"更重要。 没有边界的 PRD 一定会失控。
- 具体到可执行。 禁止模糊描述("合适的颜色"、"良好的用户体验")。颜色给 hex,组件给 TypeScript 接口,流程给 Mermaid 图。
与其他 Skill 的衔接关系
PRD 是产品决策链的第三步,且必须先于 DESIGN:
/mrd → 分析市场需求 → MRD.md
↓
/brd → 判断值不值得做 → BRD.md
↓
/prd-writing → 定义产品结构 → PRD/ 文件夹(含 03-design-handoff.md,本 skill)
↓
/design-spec → 基于 03-design-handoff.md 产出视觉规范 → DESIGN.md
↓
Claude Code 默认能力 → 基于 PRD/ + DESIGN.md 实现 MVP 代码
本 skill 的 Phase 0 会检测当前目录是否有 BRD.md / MRD.md,有就自动读取交接区数据,跳过已确认的内容。
🔒 PRD vs DESIGN 的硬性边界(最重要)
PRD 决定结构(页面/组件/路由/数据/逻辑),DESIGN 决定视觉(颜色/字体/间距/动效)。
- ✅ PRD 阶段产出
03-design-handoff.md——这是给 design-spec 的输入清单(产品调性、目标用户、参考竞品、i18n 要求、组件清单的视觉密度提示),不含任何 hex 颜色、字体 URL、CSS 变量 - ✅ design-spec 阶段读
PRD/03-design-handoff.md+04-pages-components.md,产出DESIGN.md(完整 token + 组件样式) - ❌ PRD 阶段不调用 design-spec,不产 DESIGN.md,不让用户中途跳到 design-spec
- ❌ PRD 阶段不在 03-design-handoff.md 里写具体色板/字体 URL——那是 DESIGN.md 的事
- ❌ 不读取
./DESIGN.md(即使存在也不读,因为 DESIGN 在 PRD 之后) - ❌ 不扫描
~/.claude/skills/下的设计 skill 来"借用方法论"——本 skill 只产 handoff,不产 token
为什么这样划线: 0→1 项目还没有"已存在的设计系统"作为约束,必须先定结构才能定皮肤。如果反过来(先 DESIGN 再 PRD),DESIGN 阶段不知道有哪些组件要被定规范,做出的设计是空中楼阁。这条流程顺序在教学场景里是绝对的。
参考文件(这些文件必须加载)
本 skill 有一份参考文件,生成 PRD 时必须加载:
~/.claude/skills/prd-writing/references/module-specs.md——定义 PRD 各模块的"必含字段"和"写作规范"。Phase 4 写 PRD 时按这个模板输出。
时机:进入 Phase 4 之前加载 module-specs.md。不要在 Phase 0 就加载,避免污染上下文。
旧版本的
references/style-guide.md已弃用——前端风格定调由独立的/design-specskill 在 PRD 完成后承担,不再属于 PRD 工作流。
工作流程
Phase 0:上游检测 + 判断起点
用户触发时,第一件事永远是检测上游文档。
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 668 lines · 260 tokens per session scan A 7a0515b53d5e
prd-writing is a skill published in the GitHub repository limengzhe27-boop/claude-product-doc-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 260 tokens to every session and 9,850 once invoked, about $0.0013 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
gtm-domain-library
Go-to-market domain knowledge for consumer marketplace, membership, subscription and AI-feature products: sharpening an ICP until it excludes someone, running a real user-interview panel, differentiating when the capability is rentable, launch-post anatomy, the value-metric pricing tree, the four-part deal test, and…
ai-shaped-readiness-advisor
Assess whether your product work is AI-first or AI-shaped. Use when evaluating AI maturity and choosing the next team capability to build.
agent-orchestration-advisor
Design multi-agent AI workflows with clear boundaries, handoffs, and monitoring. Use when a complex PM task should run as parallel specialized agents instead of one linear process.
company-intel
Research a company, industry, or competitor set using web search and seven analytical lenses. Use when you need structured intel that feeds downstream PM skills.
context-engineering-advisor
Diagnose context stuffing vs. context engineering. Use when an AI workflow feels bloated, brittle, or hard to steer reliably.
intelligence-collection-disciplines
Run competitive research like an intelligence agency: eight collection disciplines (OSINT to MASINT), signal-to-inference chains, and fusion. Use when one-source research isn't enough.