Token导航 LogoToken导航

Agent Skill · 研究检索

product-manager(产品经理)

product-manager 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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。

product-manager 是研究检索类 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;许可证为 MIT,暂无第三方扫描结果,按安装量在研究检索类的 73,675 个 Skill 里排第 54,403。

总安装 512周安装 22GitHub Stars 公开资料未说明下载 180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
MIT

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:product-manager(产品经理)
来源仓库:https://github.com/srstomp/pokayokay
仓库路径:skills/product-manager
安装命令:
npx skills add srstomp/pokayokay --skill "product-manager"
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。

AgentSkills.tonpx skills
npx skills add srstomp/pokayokay --skill "product-manager"

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

product-manager 用于需求分析与产品规划,支持用户故事拆解、优先级排序和价值评估。它可生成 PRD 草稿、绘制流程图或估算交付周期,辅助团队协作决策。

安装方式:npx skills add srstomp/pokayokay --skill "product-manager",需输入业务背景和目标用户画像。产出文档应基于真实数据而非假设,避免误导技术实现方向。涉及多利益方时需平衡各方诉求,明确取舍标准和验收条件。

来源仓库内路径:skills/product-manager

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 srstomp/pokayokay --skill "product-manager";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

product-manager 怎么安装?

在终端执行 npx skills add srstomp/pokayokay --skill "product-manager"。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

product-manager 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。

product-manager 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。

product-manager 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

product-manager 用的人多吗?

公开数据是累计安装 512 次,近一周安装 22 次,下载 180 次。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。

product-manager 安全吗?

没有查到第三方扫描结果,安全性无法从公开资料判断,安装前请自行审查源码、依赖和命令行为。

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。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 没有第三方扫描结果,安全性只能靠自己读源码判断。
  • 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。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 要求有安全扫描结论才肯安装的团队。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
product-manager

产品经理 · 本页

512—需要联网MIT未收录

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

name
product-manager
description
Audits feature completeness by scanning codebases and comparing against PRD requirements. Identifies gaps between backend implementation and user-facing accessibility. Generates remediation tasks and integrates with prd-analyzer output. Supports multiple frameworks including Next.js, React Router, TanStack, React Native, Expo, and more.

Product Manager

Ensures features are not just implemented but actually user-accessible. Bridges the gap between "code complete" and "user can use it."

Integrates with:

  • prd-analyzer — Reads tasks.db, features.json, PROJECT.md
  • All implementation skills — Validates their output is user-facing

The Problem This Solves

PRD says: "User can export analytics to Tableau"
Backend:   ✓ analytics-api.ts exists with Tableau endpoints
Frontend:  ✗ No /analytics route
Navigation: ✗ No menu item
Result:    Feature "done" in tasks.db, but users can't access it

Core Workflow

1. DISCOVER   → Find project structure and framework
2. READ       → Load PRD features from prd-analyzer output
3. SCAN       → Search codebase for implementation evidence
4. AUDIT      → Check user-facing accessibility
5. ANALYZE    → Compare requirements vs reality
6. REPORT     → Generate gap analysis
7. REMEDIATE  → Create tasks for missing pieces

Quick Start

1. Discover Project

# Identify framework and structure
ls -la
cat package.json

Look for:

  • Framework indicators (next.config.js, vite.config.ts, expo.json, etc.)
  • Source structure (src/, app/, pages/, etc.)
  • Backend location (backend/, server/, api/, etc.)

2. Load PRD Context

# From prd-analyzer output
cat .claude/PROJECT.md
cat .claude/features.json
sqlite3 .claude/tasks.db "SELECT id, title, status FROM epics"

3. Run Feature Audit

For each feature:

  1. Check backend implementation
  2. Check frontend implementation
  3. Check user accessibility
  4. Record findings

4. Generate Report

Output:

  • audit-report.md — Human-readable gap analysis
  • audit-results.json — Programmatic results
  • Updated tasks.db — Remediation tasks added

Feature Completeness Model

Completeness Levels

LevelNameMeaning
0Not StartedNo implementation evidence
1Backend OnlyService/API exists, no frontend
2Frontend ExistsUI components exist, not accessible
3RoutableHas route/screen, not in navigation
4AccessibleIn navigation, users can reach it
5CompleteAccessible + documented + tested

Completeness Checklist

## Feature: [Name]

### Implementation
- [ ] Backend service/API implemented
- [ ] Database schema exists (if needed)
- [ ] Frontend components exist
- [ ] API integration complete

### Accessibility  
- [ ] Route/screen exists
- [ ] Reachable from navigation
- [ ] Mobile responsive (if web)
- [ ] Works on target platforms

### Polish
- [ ] Error states handled
- [ ] Loading states present
- [ ] Empty states designed
- [ ] Documented in help/docs

### Verdict: Level [0-5]

Framework Detection

Automatic Detection

// Detection order
const frameworkIndicators = {
  // Web Frameworks
  'next.config.js':     'nextjs',
  'next.config.mjs':    'nextjs',
  'next.config.ts':     'nextjs',
  'vite.config.ts':     'vite',
  'vite.config.js':     'vite',
  'remix.config.js':    'remix',
  'astro.config.mjs':   'astro',
  
  // React Router / TanStack
  'src/routes.tsx':     'react-router',
  'src/router.tsx':     'tanstack-router',
  'app/routes/':        'remix',
  
  // Mobile
  'app.json':           'expo',
  'expo.json':          'expo',
  'react-native.config.js': 'react-native',
  'ios/':               'react-native',
  'android/':           'react-native',
  
  // Backend
  'backend/':           'separate-backend',
  'server/':            'separate-backend',
  'api/':               'api-routes',
};

Framework-Specific Patterns

FrameworkRoutes LocationNavigationAPI Calls
Next.js (pages)pages/**/*.tsxcomponents/navlib/api, services/
Next.js (app)app/**/page.tsxapp/layout.tsxapp/api/, lib/
React Routersrc/routes.tsxsrc/components/src/api/, src/services/
TanStack Routersrc/routes/src/components/src/lib/
Remixapp/routes/app/root.tsxapp/routes/*.server.ts
React Nativesrc/screens/src/navigation/src/api/, src/services/
Expo Routerapp/app/_layout.tsxsrc/api/

Detailed patterns: See references/framework-patterns.md


Scanning Process

Backend Scan

Find evidence of implementation:

# Services
find backend/src/services -name "*.ts" | head -20
grep -l "export.*class\|export.*function" backend/src/services/*.ts

# API routes/handlers
find . -path "*/api/*" -name "*.ts" | head -20
find . -path "*/handlers/*" -name "*.ts" | head -20

# Database models/schema
find . -name "schema.ts" -o -name "models.ts" -o -name "*.model.ts"

Frontend Scan

Find UI implementation:

# Routes/pages (framework-dependent)
find . -path "*/pages/*" -name "*.tsx" 2>/dev/null
find . -path "*/app/*" -name "page.tsx" 2>/dev/null
find . -path "*/screens/*" -name "*.tsx" 2>/dev/null
find . -path "*/routes/*" -name "*.tsx" 2>/dev/null

# Components
find . -path "*/components/*" -name "*.tsx" | grep -i "FEATURE_NAME"

# Navigation
grep -r "href=\|to=\|navigate\|Link" --include="*.tsx" src/components/nav/

API Integration Scan

Verify frontend calls backend:

# Find API calls
grep -r "fetch\|axios\|useMutation\|useQuery" --include="*.tsx" src/

# Find service imports
grep -r "import.*from.*services\|import.*from.*api" --include="*.tsx" src/

Navigation Scan

Check if feature is reachable:

# Find navigation components
find . -name "*nav*" -o -name "*sidebar*" -o -name "*menu*" | grep -E "\.(tsx|jsx)$"

# Check for links to feature
grep -r "/FEATURE_ROUTE" --include="*.tsx" src/

Audit Output

Audit Report Structure

# Feature Audit Report

**Project**: VoiceForm AI
**Audit Date**: 2026-01-12
**Framework**: Next.js (App Router) + Separate Backend

## Summary

| Metric | Count |
|--------|-------|
| Total Features | 30 |
| Fully Complete (L5) | 12 |
| Accessible (L4) | 5 |
| Routable (L3) | 3 |
| Frontend Exists (L2) | 2 |
| Backend Only (L1) | 8 |
| Not Started (L0) | 0 |

**Overall Completion**: 40% fully user-accessible

## Critical Gaps (P0 Features)

| Feature | Level | Missing |
|---------|-------|---------|
| F028 Analytics API | L1 | Frontend route, Navigation, UI |
| F029 Tenant Isolation | L1 | Settings UI, BYOK config screen |

## All Features

### F001: Survey Studio
**Level**: 5 - Complete ✅

**Evidence**:
- Backend: `backend/src/services/survey-studio.ts` ✓
- Route: `app/surveys/new/page.tsx` ✓
- Navigation: Sidebar "Create Survey" link ✓
- Documentation: Help article exists ✓

---

### F028: Analytics API
**Level**: 1 - Backend Only 🔴

**Evidence**:
- Backend: `backend/src/services/analytics-api.ts` ✓
- Route: ❌ No `/analytics` route found
- Navigation: ❌ No analytics link in navigation
- Documentation: ❌ No help article

**Remediation Required**:
1. Create `app/analytics/page.tsx`
2. Add Analytics to main navigation
3. Build dashboard components
4. Document analytics features

---

[...continues for all features...]

JSON Output

{
  "audit_date": "2026-01-12",
  "project": "VoiceForm AI",
  "framework": {
    "frontend": "nextjs-app",
    "backend": "separate",
    "mobile": null
  },
  "summary": {
    "total_features": 30,
    "by_level": {
      "L5_complete": 12,
      "L4_accessible": 5,
      "L3_routable": 3,
      "L2_frontend_exists": 2,
      "L1_backend_only": 8,
      "L0_not_started": 0
    }
  },
  "features": [
    {
      "id": "F028",
      "title": "Analytics API",
      "level": 1,
      "level_name": "backend_only",
      "evidence": {
        "backend": {
          "found": true,
          "files": ["backend/src/services/analytics-api.ts"]
        },
        "frontend": {
          "found": false,
          "files": []
        },
        "route": {
          "found": false,
          "path": null
        },
        "navigation": {
          "found": false,
          "location": null
        }
      },
      "remediation": [
        {
          "type": "create_route",
          "description": "Create analytics page",
          "path": "app/analytics/page.tsx"
        },
        {
          "type": "add_navigation",
          "description": "Add Analytics to sidebar"
        }
      ]
    }
  ]
}

Remediation Task Generation

Task Creation

For each gap, generate tasks in tasks.db:

-- New story for missing frontend
INSERT INTO stories (id, epic_id, title, description, status)
VALUES (
  'story-028-frontend',
  'epic-028',
  'Analytics Frontend Implementation',
  'Create user-facing analytics dashboard',
  'todo'
);

-- Tasks for the story
INSERT INTO tasks (id, story_id, title, task_type, estimate_hours, status)
VALUES 
  ('task-028-f01', 'story-028-frontend', 'Create /analytics route', 'frontend', 4, 'todo'),
  ('task-028-f02', 'story-028-frontend', 'Build analytics dashboard components', 'frontend', 8, 'todo'),
  ('task-028-f03', 'story-028-frontend', 'Add Analytics to navigation', 'frontend', 1, 'todo'),
  ('task-028-f04', 'story-028-frontend', 'Connect to analytics API', 'frontend', 4, 'todo');

Priority Assignment

Gap TypePriorityRationale
P0 feature backend-onlyP0Critical feature unusable
P1 feature backend-onlyP1Important feature unusable
Missing navigationP1Feature exists but hidden
Missing documentationP2Feature works but undiscoverable
Missing error statesP2Polish issue

Integration with prd-analyzer

Reading PRD Context

import sqlite3
import json

def load_prd_context(project_path: str):
    """Load features from prd-analyzer output"""
    
    # Load features.json
    with open(f"{project_path}/.claude/features.json") as f:
        features = json.load(f)
    
    # Load from tasks.db
    conn = sqlite3.connect(f"{project_path}/.claude/tasks.db")
    conn.row_factory = sqlite3.Row
    
    epics = conn.execute("""
        SELECT e.*, 
               COUNT(s.id) as story_count,
               SUM(CASE WHEN s.status = 'done' THEN 1 ELSE 0 END) as done_count
        FROM epics e
        LEFT JOIN stories s ON s.epic_id = e.id
        GROUP BY e.id
    """).fetchall()
    
    return {
        "features": features,
        "epics": [dict(e) for e in epics]
    }

Updating Task Status

When audit finds discrepancies:

-- Mark feature as needing frontend work
UPDATE epics 
SET status = 'needs_frontend', 
    updated_at = datetime('now')
WHERE id = 'epic-028';

-- Add audit metadata
INSERT INTO epic_metadata (epic_id, key, value)
VALUES ('epic-028', 'audit_level', '1'),
       ('epic-028', 'audit_date', '2026-01-12'),
       ('epic-028', 'audit_gaps', 'route,navigation,documentation');

Running an Audit

Full Audit Command

# In Claude Code session:

# 1. Read project structure
cat package.json
ls -la src/ app/ pages/ 2>/dev/null

# 2. Load PRD context
cat .claude/features.json
sqlite3 .claude/tasks.db "SELECT id, title, priority FROM epics ORDER BY id"

# 3. For each feature, scan for evidence
# (Claude does this systematically)

# 4. Generate report
# Outputs to .claude/audit-report.md and .claude/audit-results.json

Quick Audit (Single Feature)

# Audit just F028
# 1. Find backend
find . -name "*analytics*" -type f

# 2. Find frontend  
find . -path "*app*" -o -path "*pages*" | xargs grep -l "analytics" 2>/dev/null

# 3. Check navigation
grep -r "analytics" --include="*.tsx" src/components/nav/ app/layout.tsx

# 4. Report finding
echo "F028: Level 1 - Backend only, no frontend route"

Anti-Patterns

Audit Anti-Patterns

Anti-PatternProblemFix
Trusting tasks.db status"Done" ≠ user-facingAlways verify in codebase
Only checking file existenceFile may be empty/stubCheck for real implementation
Ignoring navigationFeature unreachableVerify menu/nav links
Skipping mobileDesktop-only isn't completeCheck responsive/native
No documentation checkUsers can't discoverVerify help/docs exist

Remediation Anti-Patterns

Anti-PatternProblemFix
Generic tasks"Add frontend" too vagueSpecific: "Create /analytics route"
Missing dependenciesFrontend before backendCheck implementation order
Overloading50 tasks at oncePrioritize by P-level
No estimatesCan't planAdd hour estimates

Checklist: Running Product Audit

Before Audit

  • [ ] prd-analyzer has run (tasks.db exists)
  • [ ] Project structure understood
  • [ ] Framework identified
  • [ ] Backend location known
  • [ ] Frontend location known

During Audit

  • [ ] Each feature checked systematically
  • [ ] Backend evidence recorded
  • [ ] Frontend evidence recorded
  • [ ] Route existence verified
  • [ ] Navigation links verified
  • [ ] Level assigned (0-5)

After Audit

  • [ ] Report generated (markdown)
  • [ ] Results saved (JSON)
  • [ ] Remediation tasks created
  • [ ] tasks.db updated
  • [ ] Priorities assigned

References:

平台分布

Claude Code

26.11%
按下载量换算47

trae

22.84%
按下载量换算41

Antigravity

16.44%
按下载量换算30

windsurf

11.78%
按下载量换算21

github-copilot

8.29%
按下载量换算15

Codex

3.7%
按下载量换算7

安全审计

暂无安全审计结果可展示。

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。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