Token导航 LogoToken导航

Agent Skill · 研究检索

roadmap-analyzer(路线图分析器)

roadmap-analyzer 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。

roadmap-analyzer 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,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 31,727。

总安装 2,646周安装 106GitHub Stars 21下载 856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:roadmap-analyzer(路线图分析器)
来源仓库:https://github.com/shipshitdev/library
仓库路径:skills/roadmap-analyzer
安装命令:
npx skills add https://github.com/shipshitdev/library --skill roadmap-analyzer
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/shipshitdev/library --skill roadmap-analyzer

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。

用于路线图分析与信息检索,适合项目规划与需求调研场景。可根据关键词快速定位候选方案或参考资料,辅助决策制定。通过 npx 命令从 shipshitdev/library 仓库安装,操作简单直接。需注意该技能可能涉及外部数据源访问,应评估网络与安全策略。

roadmap-analyzer 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/roadmap-analyzer

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/shipshitdev/library --skill roadmap-analyzer;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

roadmap-analyzer 怎么安装?

在终端执行 npx skills add https://github.com/shipshitdev/library --skill roadmap-analyzer。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

roadmap-analyzer 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。

roadmap-analyzer 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。

roadmap-analyzer 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。

roadmap-analyzer 用的人多吗?

公开数据是累计安装 2,646 次,近一周安装 106 次,下载 856 次,GitHub 21 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。

roadmap-analyzer 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 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。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 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
roadmap-analyzer

路线图分析器 · 本页

2,64621需要联网MIT通过

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

Roadmap Analyzer

Overview

This skill analyzes a project's current feature set against the needs of the Ideal Customer Profile (ICP). Before analyzing, discover the project's ICP from documentation (.agents/SYSTEM/ or project docs) and current features from the codebase and documentation.

The skill produces three outputs:

  1. Gap Analysis Table: Side-by-side comparison of ICP needs vs current features with gaps highlighted
  2. Prioritized Backlog: Ranked list of features to build next based on ICP alignment and impact
  3. Strategic Themes: High-level focus areas with specific initiatives

When to Use This Skill

Invoke this skill when asked to:

  • "What should we build next?"
  • "Compare our features against what our customers need"
  • "Analyze product gaps for our ICP"
  • "Help me plan the product roadmap"
  • "What features are missing for our target customers?"
  • "Evaluate our competitive positioning"
  • "What do we need to focus on to serve SMBs creating UGC?"

Analysis Workflow

Step 1: Discover Current Features

Use a hybrid approach to identify existing project capabilities (discover from project):

1.1 Automated Codebase Scan

Search the codebase for feature implementations. Focus on:

Key directories to explore:

apps/frontend-apps/*/      - Frontend applications
apps/apis/*/               - Backend APIs
packages/*/                - Shared packages

Search patterns for features:

  • Component names (React components often indicate UI features)
  • API endpoints (routes indicate backend capabilities)
  • Service methods (business logic indicates functionality)
  • Database schemas (data models reveal supported features)
  • Configuration files (feature flags, settings)

Use grep/glob to find:

  • Video editing features: pattern: "edit|timeline|trim|cut|transition"
  • Avatar features: pattern: "avatar|character|persona|voice"
  • Export features: pattern: "export|format|platform|render"
  • Template features: pattern: "template|preset|brand"
  • AI features: pattern: "ai|generate|auto|openai|anthropic"
1.2 User Context Integration

Ask the user:

  • "Are there features I should know about that might not be obvious in the code?"
  • "Any recent features added that might not be fully integrated?"
  • "Features in beta or soft-launched?"
  • "Integrations or capabilities documented elsewhere?"
1.3 Feature Inventory Creation

Organize discovered features into categories from references/feature-categories.md:

  1. Content Input & Ideation
  2. AI Avatar Creation & Management
  3. Video Editing & Assembly
  4. Visual Enhancement & Effects
  5. Text, Captions & Typography
  6. Audio & Music
  7. Platform Optimization & Export
  8. Templates & Brand Management
  9. Collaboration & Workflow
  10. Asset Management & Media Library
  11. AI & Automation Features
  12. Analytics & Performance
  13. Learning & Support
  14. Integration & Ecosystem
  15. E-commerce & Monetization

For each feature, note:

  • Status: Exists (production), Partial (incomplete), Planned, Missing
  • Quality: Production-ready, Beta, Prototype, Needs improvement
  • Location: Where found in codebase (file paths)

Step 2: Load ICP and Editor Needs Context

Read the reference documents to understand requirements:

Required reading:

  • references/icp-profile.md - Deep understanding of target customer needs
  • references/editor-needs.md - What editors need for great social content
  • references/feature-categories.md - Comprehensive feature taxonomy

Key focus areas from ICP:

  1. AI Avatar Army: Creating multiple avatars, consistency, customization
  2. User Empowerment: Self-service creation, templates, intuitive UX
  3. UGC at Scale: Volume, speed, platform-native content
  4. Team Collaboration: Multi-user, approvals, brand consistency

Critical editor needs:

  • Platform-specific optimization (TikTok, Instagram, YouTube)
  • Fast pacing and trending effects
  • Text overlays and captions
  • Audio/music integration
  • Quick variations for A/B testing

Step 3: Gap Analysis

Compare current features against ICP needs and editor requirements.

3.1 Create Gap Analysis Table

Format:

| Feature Category | ICP Need Priority | Current State | Gap Severity | Key Missing Elements |
|-----------------|-------------------|---------------|--------------|----------------------|
| AI Avatar Creation | CRITICAL | Partial | HIGH | Multiple avatars, avatar library, voice cloning |
| Platform Optimization | CRITICAL | Missing | CRITICAL | Auto-reframe, batch export multiple formats |
| ... | ... | ... | ... | ... |

Gap Severity Levels:

  • CRITICAL: Blocking ICP adoption, core value prop missing
  • HIGH: Significantly impacts ICP success, competitive disadvantage
  • MEDIUM: Limits scale/efficiency, nice-to-have
  • LOW: Future enhancement, not blocking

Priority Levels:

  • CRITICAL: Must-have for ICP, core to value proposition
  • HIGH: Important for ICP success, competitive requirement
  • MEDIUM: Valuable but not essential
  • LOW: Nice-to-have, future consideration
3.2 Identify Critical Gaps

Highlight gaps where:

  1. ICP Need Priority = CRITICAL AND Gap Severity = CRITICAL/HIGH
  2. Feature is table stakes for competitors
  3. Feature directly enables "AI avatar army" or "user empowerment"
  4. Feature required for platform-native content (TikTok, Instagram, etc.)

Step 4: Generate Prioritized Backlog

Create a ranked list of features to build next.

4.1 Prioritization Framework

Score each feature on:

ICP Impact (1-5)

  • 5: Core to ICP value prop, directly addresses primary pain point
  • 4: Strongly supports ICP needs, key differentiator
  • 3: Valuable for ICP, improves experience
  • 2: Nice-to-have for ICP
  • 1: Minimal ICP impact

Urgency (1-5)

  • 5: Blocking sales/adoption, critical competitive gap
  • 4: Significant competitive pressure, customer requests
  • 3: On roadmap, good timing to build
  • 2: Can wait, not time-sensitive
  • 1: Future consideration

Implementation Effort (1-5)

  • 5: Very high effort, 3+ months, complex
  • 4: High effort, 1-3 months, significant work
  • 3: Medium effort, 2-4 weeks, moderate complexity
  • 2: Low effort, < 2 weeks, straightforward
  • 1: Very low effort, < 1 week, simple

Priority Score = (ICP Impact × 2 + Urgency × 1.5) / Implementation Effort

Higher score = higher priority.

4.2 Backlog Format
## Prioritized Feature Backlog

### P0 (Build Immediately - Priority Score > 3.0)
1. **[Feature Name]** (Score: 4.2)
   - **Why**: [ICP impact explanation]
   - **Customer need**: [Specific ICP pain point addressed]
   - **Effort**: [Time estimate]
   - **Dependencies**: [What's needed first]

### P1 (Build Next Quarter - Priority Score 2.0-3.0)
[Same format]

### P2 (Future Consideration - Priority Score < 2.0)
[Same format]
4.3 Quick Wins Section

Identify features with:

  • High ICP Impact (4-5)
  • Low Implementation Effort (1-2)
  • Decent Urgency (3+)

These are "quick wins" that deliver high value with low investment.

Step 5: Define Strategic Themes

Group features into high-level strategic focus areas.

5.1 Theme Identification

Analyze feature clusters to identify 3-5 strategic themes, such as:

Example themes:

  • "AI Avatar Arsenal": Building the avatar creation and management system
  • "Platform-Native Mastery": Optimizing for TikTok, Instagram, YouTube
  • "Team Empowerment Engine": Collaboration, templates, workflows
  • "Speed to Market": Automation, batch processing, quick variations
  • "Editor's Toolkit": Core editing capabilities for creators
5.2 Theme Structure

For each theme, provide:

## [Theme Name]

**Vision**: [1-2 sentence vision for this theme]

**ICP Alignment**: [How this theme serves ICP needs]

**Current State**: [Where we are today]

**Target State**: [Where we need to be]

**Key Initiatives**:
1. **[Initiative Name]**
   - Features: [Specific features]
   - Impact: [Expected outcome]
   - Timeline: [When to build]

2. **[Next initiative]**
   [Same format]

**Success Metrics**:
- [Metric 1]: [Target]
- [Metric 2]: [Target]
5.3 Theme Prioritization

Recommend theme focus order based on:

  1. Foundation themes first: Core platform capabilities required for everything else
  2. Differentiator themes next: Unique value props (e.g., AI avatar army)
  3. Enhancement themes last: Improvements to existing capabilities

Step 6: Output Assembly

Combine all analysis into a comprehensive report:

# [Project] Roadmap Analysis
## [Date]

## Executive Summary
[3-5 bullets summarizing key findings and recommendations]

## Gap Analysis
[Gap analysis table from Step 3]

## Critical Gaps
[Detailed explanation of critical gaps from Step 3.2]

## Prioritized Backlog
[Prioritized backlog from Step 4]

## Quick Wins
[Quick wins list from Step 4.3]

## Strategic Themes
[Strategic themes from Step 5]

## Recommended Focus
**Next 30 days**: [Immediate priorities]
**Next quarter**: [Q1 focus areas]
**Next 6 months**: [Half-year vision]

## Appendix
### Feature Inventory
[Complete list of discovered features]

### ICP Summary
[Brief ICP recap]

Best Practices

Analysis Quality

  1. Be specific: Don't just say "add editing features"—specify "Add timeline trimming with frame-level precision"
  2. Show evidence: Reference where features exist (or don't) in codebase
  3. Connect to customer: Explain how each feature serves ICP needs
  4. Consider competition: Note when gaps create competitive disadvantage
  5. Balance ambition with reality: Acknowledge technical constraints

Prioritization Rigor

  1. Defend scores: Explain why a feature gets specific impact/urgency/effort scores
  2. Consider dependencies: Note when features require others to be built first
  3. Think sequentially: Some features enable others (e.g., avatar library before multi-avatar videos)
  4. Question assumptions: If effort seems high, suggest simpler MVP versions

Strategic Thinking

  1. Theme coherence: Themes should feel cohesive, not arbitrary groupings
  2. ICP-centric: Every recommendation should trace back to ICP needs
  3. Differentiation: Emphasize features that set the project apart
  4. Reality check: Acknowledge market timing, competitive landscape, technical feasibility

Common Pitfalls to Avoid

  1. Don't just list features: Explain WHY each feature matters for ICP
  2. Don't ignore existing features: Give credit for what already exists
  3. Don't treat all gaps equally: Not all missing features are critical
  4. Don't forget the "AI avatar army" vision: This is a key differentiator
  5. Don't overlook platform-specific needs: TikTok ≠ YouTube
  6. Don't assume features work well: Existing features might have quality gaps

Resources

references/

Three comprehensive reference documents inform the analysis:

  • icp-profile.md: Detailed profile of target customers (SMBs creating UGC with AI, empowering users, building AI avatar armies). Includes pain points, core needs, success metrics, and buying journey.
  • editor-needs.md: Comprehensive breakdown of what video editors need to create great content for social platforms (TikTok, Instagram, YouTube). Covers editing capabilities, platform-specific requirements, audio/visual needs, and competitive benchmarks.
  • feature-categories.md: Taxonomy of 15 feature categories typical in AI UGC platforms. Provides framework for organizing discovered features and identifying gaps systematically.

Load these references at the start of analysis to inform all evaluation and recommendations.

平台分布

Claude Code

30.92%
按下载量换算265

Gemini CLI

25.33%
按下载量换算217

Antigravity

16.66%
按下载量换算143

OpenCode

13.25%
按下载量换算113

Codex

8.08%
按下载量换算69

Cursor

3.87%
按下载量换算33

安全审计

Gen Agent Trust Hub

通过

Socket

通过

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。

继续浏览同类 Skills