Token导航 LogoToken导航

Agent Skill · 前端设计

pm-prd-write(pm prd 写)

pm-prd-write 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

pm-prd-write 是前端设计类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在前端设计类的 10,216 个 Skill 里排第 9,864。

总安装 192周安装 8GitHub Stars 11下载 64核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:pm-prd-write(pm prd 写)
来源仓库:https://github.com/parhumm/jaan-to
仓库路径:skills/pm-prd-write
安装命令:
npx skills add https://github.com/parhumm/jaan-to --skill pm-prd-write
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/parhumm/jaan-to --skill pm-prd-write

这个 Skill 是做什么的

前端设计类 Skill 关注页面结构、视觉规范和组件写法,输出通常是可运行的界面代码。

pm-prd-write 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态或协作事项进行整理。适用于需要分析代码变更、跟踪 Issue 进展或管理 Pull Request 的 AI 开发场景。

通过 npx skills add 命令从指定 GitHub 仓库安装,需确认权限范围和文件操作权限。使用前建议检查仓库维护状态,避免触发不必要的网络请求或文件读写操作。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/pm-prd-write

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/parhumm/jaan-to --skill pm-prd-write;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

pm-prd-write 怎么安装?

在终端执行 npx skills add https://github.com/parhumm/jaan-to --skill pm-prd-write。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

pm-prd-write 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

pm-prd-write 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

pm-prd-write 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

pm-prd-write 用的人多吗?

公开数据是累计安装 192 次,近一周安装 8 次,下载 64 次,GitHub 11 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

pm-prd-write 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理前端设计类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

前端设计类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
frontend-design

前端设计 · Codex、Claude

8,823,360126,072只读MIT通过
vercel-composition-patterns

Vercel 组合模式 · Codex、Claude

3,858,81625,893需要联网MIT通过
vercel-react-native-skills

Vercel React native skills 前端 · Codex、Claude

2,468,66425,878只读MIT通过
adapt

adapt 搜索 · Codex、Claude

2,027,04023,125操作浏览器未明确通过
bolder

bolder 分析 · Codex、Claude

2,004,28823,130只读未明确通过
pm-prd-write

pm prd 写 · 本页

19211需要联网未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

pm-prd-write

Generate a PRD from initiative description.

Context Files

  • $JAAN_CONTEXT_DIR/config.md - Configuration
  • $JAAN_CONTEXT_DIR/boundaries.md - Trust rules
  • $JAAN_TEMPLATES_DIR/jaan-to-pm-prd-write.template.md - PRD template
  • $JAAN_LEARN_DIR/jaan-to-pm-prd-write.learn.md - Past lessons (loaded in Pre-Execution)
  • $JAAN_CONTEXT_DIR/tech.md - Tech context (if exists)
  • $JAAN_CONTEXT_DIR/team.md - Team context (if exists)
  • ${CLAUDE_PLUGIN_ROOT}/docs/extending/language-protocol.md - Language resolution protocol

Input

Initiative: $ARGUMENTS

IMPORTANT: The initiative above is your input. Use it directly. Do NOT ask for the initiative again.


Pre-Execution Protocol

MANDATORY — Read and execute ALL steps in: ${CLAUDE_PLUGIN_ROOT}/docs/extending/pre-execution-protocol.md Skill name: pm-prd-write Execute: Step 0 (Init Guard) → A (Load Lessons) → B (Resolve Template) → C (Offer Template Seeding)

Also read context files if available:

  • $JAAN_CONTEXT_DIR/tech.md - Know the tech stack to reference
  • $JAAN_CONTEXT_DIR/team.md - Know team structure and norms

If the file does not exist, continue without it.

Language Settings

Read and apply language protocol: ${CLAUDE_PLUGIN_ROOT}/docs/extending/language-protocol.md Override field for this skill: language_pm-prd-write


PHASE 1: Analysis (Read-Only)

Step 1: Gather Information

You already have the initiative from the Input section above. Now ask these questions:

  1. "What problem does this solve for users?"
  2. "How will you measure success? (specific metrics)"
  3. "What is explicitly NOT included in this scope?"
  4. "Do you have design images, screenshots, or visual references to include? If yes, provide file/folder paths."

Step 2: Plan PRD Structure

After receiving answers, mentally outline:

  • Title (from initiative)
  • Problem Statement (from Q1)
  • Success Metrics (from Q2)
  • Scope boundaries (from Q3)
  • User Stories (derive 3+ from context)

HARD STOP - Human Review Check

Before generating the PRD, confirm with the user:

"I have all the information needed. Ready to generate the PRD for '{initiative}'? [y/n]"

Do NOT proceed to Phase 2 without explicit approval.


PHASE 2: Generation (Write Phase)

Step 3: Generate PRD

Use the template from: $JAAN_TEMPLATES_DIR/jaan-to-pm-prd-write.template.md

Fill all sections:

  • Title: From initiative
  • Problem Statement: From question 1
  • Solution Overview: Synthesized from input
  • Success Metrics: From question 2 (table format, minimum 2)
  • Scope: In-scope from input, out-of-scope from question 3
  • User Stories: Derived from problem + solution (minimum 3)
  • Open Questions: Any unresolved items

Image Embedding

If user provided image/screenshot paths in Step 1 Q4, embed them inline using ![description](resolved-path) in relevant sections (Solution Overview, User Stories, Appendix). Use URL-encoded paths for spaces/special characters.

Tech Context Integration

If $JAAN_CONTEXT_DIR/tech.md exists:

  1. Read the file to understand tech stack
  2. When generating User Stories:

- Reference appropriate backend framework (e.g., "API endpoint in FastAPI") - Reference appropriate frontend framework (e.g., "React component with Redux state") - Reference mobile platforms if applicable

  1. When generating Technical Constraints:

- Include relevant constraints from tech.md - Reference specific technical requirements (e.g., "Must maintain sub-200ms p95 latency")

  1. When generating Success Metrics:

- Consider performance targets from tech.md - Include technical success criteria where relevant

If tech.md doesn't exist:

  • Generate generic technical references
  • Add note in PRD: "Update $JAAN_CONTEXT_DIR/tech.md for tech-aware PRDs"

Step 4: Quality Check

Before showing preview, verify:

  • Has clear problem statement
  • Has at least 2 measurable success metrics
  • Has explicit out-of-scope section
  • Has at least 3 user stories
  • If user provided images: all references use ![alt](...) syntax with URL-encoded paths

If any check fails, revise before preview.

Step 5: Preview & Approval

Show the complete PRD and ask:

"Here's the PRD preview. Approve writing to output? [y/n]"

Step 5.5: Generate ID and Folder Structure

If approved, set up the output structure:

  1. Source ID generator utility:
source "${CLAUDE_PLUGIN_ROOT}/scripts/lib/id-generator.sh"
  1. Generate sequential ID and output paths:
# Define subdomain directory
SUBDOMAIN_DIR="$JAAN_OUTPUTS_DIR/pm/prd"
mkdir -p "$SUBDOMAIN_DIR"

# Generate next ID
NEXT_ID=$(generate_next_id "$SUBDOMAIN_DIR")

# Create folder and file paths
slug="{lowercase-hyphenated-from-title-max-50-chars}"
OUTPUT_FOLDER="${SUBDOMAIN_DIR}/${NEXT_ID}-${slug}"
MAIN_FILE="${OUTPUT_FOLDER}/${NEXT_ID}-prd-${slug}.md"
  1. Preview output configuration:
Output Configuration - ID: {NEXT_ID} - Folder: $JAAN_OUTPUTS_DIR/pm/prd/{NEXT_ID}-{slug}/ - Main file: {NEXT_ID}-prd-{slug}.md

Step 5.7: Resolve & Copy Assets

If user provided image paths in Step 1 Q4:

Reference: See ${CLAUDE_PLUGIN_ROOT}/docs/extending/asset-embedding-reference.md for the asset resolution protocol (path detection, copy rules, markdown embedding).

Source ${CLAUDE_PLUGIN_ROOT}/scripts/lib/asset-handler.sh. For each image path: check is_jaan_path — if inside $JAAN_*, reference in-place; if external, ask user "Copy these external assets into output folder? [y/n]" and call copy_external_assets if approved. Use resolve_asset_path to get markdown-relative paths for the generated PRD.

Step 6: Write Output

  1. Create output folder:
mkdir -p "$OUTPUT_FOLDER"
  1. Write PRD to main file:
cat > "$MAIN_FILE" <<'EOF'
{generated PRD content with Executive Summary}
EOF
  1. Update subdomain index:
source "${CLAUDE_PLUGIN_ROOT}/scripts/lib/index-updater.sh"
add_to_index \
  "$SUBDOMAIN_DIR/README.md" \
  "$NEXT_ID" \
  "${NEXT_ID}-${slug}" \
  "{PRD Title}" \
  "{1-2 sentence executive summary from PRD}"
  1. Confirm completion:
✓ PRD written to: $JAAN_OUTPUTS_DIR/pm/prd/{NEXT_ID}-{slug}/{NEXT_ID}-prd-{slug}.md ✓ Index updated: $JAAN_OUTPUTS_DIR/pm/prd/README.md

Step 7: Auto-Invoke User Story Generation

Use AskUserQuestion:

  • Question: "Generate detailed user stories from this PRD?"
  • Header: "Stories"
  • Options:

- "Yes" — Choose stories to expand - "No" — Skip, done with PRD only

If "No", skip to Step 8.

If "Yes":

List the user stories from the generated PRD:

"Which stories to expand into full user stories?" [1] As a {persona}, I want {action} so that {benefit} [2] As a {persona}, I want {action} so that {benefit} [3] As a {persona}, I want {action} so that {benefit} [All] Generate all stories Enter numbers (e.g., "1,3" or "all"):

For each selected story, run: /jaan-to:pm-story-write "{story_statement}"

This invokes the full story-write skill with INVEST validation, Gherkin acceptance criteria, and edge case mapping.

Step 8: Capture Feedback

After PRD is written, ask:

"Any feedback or improvements needed? [y/n]"

If yes:

  1. Ask: "What should be improved?"
  2. Offer options: "How should I handle this? [1] Fix now - Update this PRD [2] Learn - Save for future PRDs [3] Both - Fix now AND save lesson"

Option 1 - Fix now:

  • Apply the feedback to the current PRD
  • Re-run Step 5 (Preview & Approval) with updated content
  • Write the updated PRD

Option 2 - Learn for future:

  • Run: /jaan-to:learn-add pm-prd-write "{feedback}"
  • Follow /jaan-to:learn-add workflow (categorize → preview → commit)

Option 3 - Both:

  • First: Apply fix to current PRD (Option 1)
  • Then: Run /jaan-to:learn-add (Option 2)

If no:

  • PRD workflow complete

Skill Alignment

  • Two-phase workflow with HARD STOP for human approval
  • Template-driven output structure
  • Generic across industries and domains
  • Output to standardized $JAAN_OUTPUTS_DIR path

Definition of Done

  • PRD file exists at correct path
  • All quality checks pass
  • User has approved the content
  • User stories generated via /jaan-to:pm-story-write (if selected)

平台分布

Codex

35.7%
按下载量换算23

Claude

29.8%
按下载量换算19

Cursor

18.45%
按下载量换算12

Gemini CLI

9.03%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

可疑

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills