Token导航 LogoToken导航

Agent Skill · 研究检索

deep-reading-analyst(深度阅读分析师)

deep-reading-analyst 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。

deep-reading-analyst 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,权限为只读;许可证为 MIT,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 23,498。

总安装 4,253周安装 179GitHub Stars 37下载 1,489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:deep-reading-analyst(深度阅读分析师)
来源仓库:https://github.com/ovachiever/droid-tings
仓库路径:skills/deep-reading-analyst
安装命令:
npx skills add https://github.com/ovachiever/droid-tings --skill deep-reading-analyst
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/ovachiever/droid-tings --skill deep-reading-analyst

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。

用于将浅层阅读转化为深度学习,应用十种以上思维框架系统分析。从 SCQA、5W2H 到心智模型,提供不同层级的分析模板。输出结构化笔记与行动建议,支持从理解到应用的闭环。时间 configurable,适用于个人学习或团队知识沉淀。

deep-reading-analyst 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/deep-reading-analyst

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/ovachiever/droid-tings --skill deep-reading-an…;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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

deep-reading-analyst 怎么安装?

在终端执行 npx skills add https://github.com/ovachiever/droid-tings --skill deep-reading-analyst。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

deep-reading-analyst 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。

deep-reading-analyst 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

deep-reading-analyst 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。

deep-reading-analyst 用的人多吗?

公开数据是累计安装 4,253 次,近一周安装 179 次,下载 1,489 次,GitHub 37 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。

deep-reading-analyst 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。
  • 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。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 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。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

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
deep-reading-analyst

深度阅读分析师 · 本页

4,25337只读MIT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

Deep Reading Analyst

Transforms surface-level reading into deep learning through systematic analysis using 10+ proven thinking frameworks. Guides users from understanding to application through structured workflows.

Framework Arsenal

Quick Analysis (15min)

  • 📋 SCQA - Structure thinking (Situation-Complication-Question-Answer)
  • 🔍 5W2H - Completeness check (What, Why, Who, When, Where, How, How much)

Standard Analysis (30min)

  • 🎯 Critical Thinking - Argument evaluation
  • 🔄 Inversion Thinking - Risk identification

Deep Analysis (60min)

  • 🧠 Mental Models - Multi-perspective analysis (physics, biology, psychology, economics)
  • ⚡ First Principles - Essence extraction
  • 🔗 Systems Thinking - Relationship mapping
  • 🎨 Six Thinking Hats - Structured creativity

Research Analysis (120min+)

  • 📊 Cross-Source Comparison - Multi-article synthesis

Workflow Decision Tree

User provides content
    ↓
Ask: Purpose + Depth Level + Preferred Frameworks
    ↓
┌─────────────────┬─────────────────┬─────────────────┬─────────────────┐
│   Level 1       │   Level 2       │   Level 3       │   Level 4       │
│   Quick         │   Standard      │   Deep          │   Research      │
│   15min         │   30min         │   60min         │   120min+       │
├─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ • SCQA          │ Level 1 +       │ Level 2 +       │ Level 3 +       │
│ • 5W2H          │ • Critical      │ • Mental Models │ • Cross-source  │
│ • Structure     │ • Inversion     │ • First Princ.  │ • Web search    │
│                 │                 │ • Systems       │ • Synthesis     │
│                 │                 │ • Six Hats      │                 │
└─────────────────┴─────────────────┴─────────────────┴─────────────────┘

Step 1: Initialize Analysis

Ask User (conversationally):

  1. "What's your main goal for reading this?"

- Problem-solving / Learning / Writing / Decision-making / Curiosity

  1. "How deep do you want to go?"

- Quick (15min) / Standard (30min) / Deep (60min) / Research (120min+)

  1. "Any specific frameworks you'd like to use?"

- Suggest based on content type (see Framework Selection Guide below)

Default if no response: Level 2 (Standard mode) with auto-selected frameworks

Framework Selection Guide

Based on content type, auto-suggest:

📄 Strategy/Business articles → SCQA + Mental Models + Inversion
📊 Research papers → 5W2H + Critical Thinking + Systems Thinking
💡 How-to guides → SCQA + 5W2H + First Principles
🎯 Opinion pieces → Critical Thinking + Inversion + Six Hats
📈 Case studies → SCQA + Mental Models + Systems Thinking

Step 2: Structural Understanding

Always start here regardless of depth level.

Phase 2A: Basic Structure

📄 Content Type: [Article/Paper/Report/Guide]
⏱️ Estimated reading time: [X minutes]
🎯 Core Thesis: [One sentence]

Structure Overview:
├─ Main Argument 1
│   ├─ Supporting point 1.1
│   └─ Supporting point 1.2
├─ Main Argument 2
└─ Main Argument 3

Key Concepts: [3-5 terms with brief definitions]

Phase 2B: SCQA Analysis (Quick Framework)

Load references/scqa_framework.md and apply:

## SCQA Structure

**S (Situation)**: [Background/context the article establishes]
**C (Complication)**: [Problem/challenge identified]
**Q (Question)**: [Core question being addressed]
**A (Answer)**: [Main solution/conclusion]

📊 Structure Quality:
- Clarity: [★★★★☆]
- Logic flow: [★★★★★]
- Completeness: [★★★☆☆]

Phase 2C: 5W2H Completeness Check (if Level 1+)

Quick scan using references/5w2h_analysis.md:

## Information Completeness

✅ Well-covered: [What, Why, How]
⚠️  Partially covered: [Who, When]
❌ Missing: [Where, How much]

🔴 Critical gaps: [List 1-2 most important missing pieces]

Step 3: Apply Thinking Models

Select based on depth level and user preference:

Level 1 (Quick - 15 min)

Core: Structure + SCQA + 5W2H Quick Check

Output:

  • SCQA breakdown
  • Information gaps (from 5W2H)
  • TOP 3 insights
  • 1 immediate action item

Level 2 (Standard - 30 min)

Add: Critical Thinking + Inversion

Load and apply:

  • references/critical_thinking.md:

- Argument quality assessment - Logic flaw identification - Evidence evaluation - Alternative perspectives

  • references/inversion_thinking.md:

- How to ensure failure? (reverse the advice) - What assumptions if wrong? - Missing risks - Pre-mortem analysis

## Critical Analysis

### Argument Strength: [X/10]
Strengths:
- [Point 1]

Weaknesses:
- [Point 1]

Logical fallacies detected:
- [If any]

## Inversion Analysis

🚨 How this could fail:
1. [Failure mode 1] → Mitigation: [...]
2. [Failure mode 2] → Mitigation: [...]

Missing risk factors:
- [Risk 1]

Level 3 (Deep - 60 min)

Add: Mental Models + First Principles + Systems + Six Hats

Load and apply:

  • references/mental_models.md:

- Select 3-5 relevant models from different disciplines - Apply each lens to the content - Identify cross-model insights

  • references/first_principles.md:

- Strip to fundamental truths - Identify core assumptions - Rebuild understanding from base

  • references/systems_thinking.md:

- Map relationships and feedback loops - Identify leverage points - See the big picture

  • references/six_hats.md:

- White (facts), Red (feelings), Black (caution) - Yellow (benefits), Green (creativity), Blue (process)

## Multi-Model Analysis

### Mental Models Applied:
1. **[Model 1 from X discipline]**
   Insight: [...]

2. **[Model 2 from Y discipline]**
   Insight: [...]

3. **[Model 3 from Z discipline]**
   Insight: [...]

Cross-model pattern: [Key insight from combining models]

### First Principles Breakdown:
Core assumptions:
1. [Assumption 1] → Valid: [Yes/No/Conditional]
2. [Assumption 2] → Valid: [Yes/No/Conditional]

Fundamental truth: [What remains after stripping assumptions]

### Systems Map:

[Variable A] ──reinforces──> [Variable B] ↑ | | | balances reinforces | | └─────────<────────────────┘

Leverage point: [Where small change = big impact]

### Six Hats Perspective:
🤍 Facts: [Objective data]
❤️ Feelings: [Intuitive response]
🖤 Cautions: [Risks and downsides]
💛 Benefits: [Positive aspects]
💚 Ideas: [Creative alternatives]
💙 Process: [Meta-thinking]

Level 4 (Research - 120 min+)

Add: Cross-source comparison via web_search

Use web_search to find 2-3 related sources, then:

  • Load references/comparison_matrix.md
  • Compare SCQA across sources
  • Identify consensus vs. divergence
  • Synthesize integrated perspective
## Multi-Source Analysis

### Source 1: [This article]
S-C-Q-A: [Summary]
Key claim: [...]

### Source 2: [Found article]
S-C-Q-A: [Summary]
Key claim: [...]

### Source 3: [Found article]
S-C-Q-A: [Summary]
Key claim: [...]

## Synthesis

**Consensus**: [What all agree on]
**Divergence**: [Where they differ]
**Unique value**: [What each contributes]
**Integrated view**: [Your synthesis]

Step 4: Synthesis & Output

Generate based on user goal:

For Problem-Solving:

## Applicable Solutions
[Extract 2-3 methods from content]

## Application Plan
Problem: [User's specific issue]
Relevant insights: [From analysis]

Action steps:
1. [Concrete action with timeline]
2. [Concrete action with timeline]
3. [Concrete action with timeline]

Success metrics: [How to measure]

## Risk Mitigation (from Inversion)
Potential failure points:
- [Point 1] → Prevent by: [...]
- [Point 2] → Prevent by: [...]

For Learning:

## Learning Notes

Core concepts (explained simply):
1. **[Concept 1]**: [Definition + Example]
2. **[Concept 2]**: [Definition + Example]

Mental models gained:
- [Model 1]: [How it works]

Connections to prior knowledge:
- [Link to something user already knows]

## Deeper Understanding (First Principles)
Fundamental question: [...]
Core principle: [...]

## Verification Questions
1. [Question to test understanding]
2. [Question to test application]
3. [Question to test evaluation]

For Writing Reference:

## Key Arguments & Evidence
[Structured extraction with page/paragraph numbers]

## Quotable Insights
"[Quote 1]" — Context: [...]
"[Quote 2]" — Context: [...]

## Critical Analysis Notes
Strengths: [For citing]
Limitations: [For balanced discussion]

## Alternative Perspectives (from Mental Models)
[What other disciplines would say about this]

## Gaps & Counterfactuals
What the article doesn't address:
- [Gap 1]
- [Gap 2]

For Decision-Making:

## Decision Framework

Options presented: [A / B / C]

Multi-model evaluation:
- Economic lens: [...]
- Risk lens (Inversion): [...]
- Systems lens: [...]

## Six Hats Decision Analysis
🤍 Facts: [Objective comparison]
🖤 Risks: [What could go wrong]
💛 Benefits: [Upside potential]
💚 Alternatives: [Other options not considered]
💙 Recommendation: [Synthesized advice]

## Scenario Analysis (from Inversion)
Best case: [...]
Worst case: [...]
Most likely: [...]

Step 5: Knowledge Activation

Always end with:

## 🎯 Immediate Takeaways (Top 3)

1. **[Insight 1]**
   Why it matters: [Personal relevance]
   One action: [Specific, time-bound]

2. **[Insight 2]**
   Why it matters: [Personal relevance]
   One action: [Specific, time-bound]

3. **[Insight 3]**
   Why it matters: [Personal relevance]
   One action: [Specific, time-bound]

## 💡 Quick Win
[One thing to try in next 24 hours - make it TINY and SPECIFIC]

## 🔗 Next Steps

**To deepen understanding:**
[ ] Further reading: [If relevant]
[ ] Apply framework X to topic Y
[ ] Discuss with: [Who could add perspective]

**To apply:**
[ ] Experiment: [Test in real context]
[ ] Teach: [Explain to someone else]
[ ] Combine: [Mix with another idea]

## 🧭 Thinking Models Used
[Checkboxes showing which frameworks were applied]
✅ SCQA ✅ 5W2H ✅ Critical Thinking ✅ Inversion
□ Mental Models □ First Principles □ Systems □ Six Hats

Quality Standards

Every analysis must:

  • ✅ Stay faithful to original content (no misrepresentation)
  • ✅ Distinguish facts from opinions
  • ✅ Provide concrete examples
  • ✅ Apply frameworks appropriately (not force-fit)
  • ✅ Connect to user's context when possible
  • ✅ End with actionable steps
  • ✅ Cite specific sections (paragraph numbers, quotes)

Avoid:

  • ❌ Overwhelming with all frameworks at once (respect depth level)
  • ❌ Academic jargon without explanation
  • ❌ Analysis without application
  • ❌ Copying text verbatim (always reword for understanding)
  • ❌ Using frameworks superficially (go deep, not wide)

Interaction Patterns

Progressive questioning:

  • Understanding: "What do you think the author means by X?"
  • Critical: "Do you see any gaps in this argument?"
  • Application: "How might you use this in your work?"
  • Meta: "Which thinking model helped you most? Why?"

Adapt to signals:

  • User asks "what's the main point?" → They want conciseness, use SCQA
  • User challenges your analysis → Lean into Critical Thinking + Inversion
  • User asks "how do I use this?" → Focus on application + First Principles
  • User wants "multiple perspectives" → Use Six Hats or Mental Models
  • User mentions "risks" → Apply Inversion Thinking
  • User asks "how does this connect?" → Use Systems Thinking

Framework suggestions during conversation:

  • "Would you like me to apply [X framework] to this point?"
  • "This seems like a good place for inversion thinking - want to explore failure modes?"
  • "I notice several mental models at play here, want me to unpack them?"

Reference Materials

Core Frameworks (All Levels)

  • references/scqa_framework.md - Structure thinking (S-C-Q-A)
  • references/5w2h_analysis.md - Completeness check (7 questions)

Standard Level Frameworks

  • references/critical_thinking.md - Argument analysis
  • references/inversion_thinking.md - Risk and failure mode analysis

Deep Level Frameworks

  • references/mental_models.md - Multi-discipline model library
  • references/first_principles.md - Essence extraction method
  • references/systems_thinking.md - Relationship mapping
  • references/six_hats.md - Multi-perspective protocol

Output Formats

  • references/output_templates.md - Note format examples
  • references/comparison_matrix.md - Cross-article analysis

Advanced Usage

Custom Framework Combinations

User can request specific combinations:

  • "Use SCQA + Inversion" - Structure with risk analysis
  • "Apply Mental Models + Systems Thinking" - Multi-lens system analysis
  • "5W2H + Critical Thinking" - Completeness + quality check

Iterative Deepening

Start with Level 1, then ask:

  • "Want to go deeper on any part?"
  • "Which framework would be most valuable here?"
  • "Should we do an inversion analysis of this solution?"

Domain-Specific Optimizations

Business/Strategy: SCQA + Mental Models (economics) + Inversion Technical/Research: 5W2H + First Principles + Critical Thinking Personal Development: Six Hats + Inversion + Systems Decision-Making: Mental Models + Inversion + SCQA Creative: Six Hats + First Principles + Mental Models


Remember: The goal is insight, not framework completion. Use frameworks as tools to reveal understanding, not as checklists to complete. Quality of thinking > quantity of frameworks applied.

平台分布

Claude Code

29.86%
按下载量换算445

Antigravity

23.71%
按下载量换算353

Gemini CLI

19.58%
按下载量换算292

OpenCode

14.43%
按下载量换算215

Cursor

7.56%
按下载量换算113

Codex

3.69%
按下载量换算55

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。

继续浏览同类 Skills