MCPInstall

deep-reading-analyst

Comprehensive framework for deep analysis of articles, papers, and long-form content using 10+ thinking models (SCQA, 5W2H, critical thinking, inversion, mental models, first principles, systems thinking, six thinking hats). Use when users want to: (1) deeply understand complex articles/content, (2) analyze arguments and identify logical flaws, (3) extract actionable insights from reading materials, (4) create study notes or learning summaries, (5) compare multiple sources, (6) transform knowledge into practical applications, or (7) apply specific thinking frameworks. Triggered by phrases like 'analyze this article,' 'help me understand,' 'deep dive into,' 'extract insights from,' 'use [framework name],' or when users provide URLs/long-form content for analysis.

Installs
75
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/yyh211/claude-meta-skill --skill deep-reading-analyst
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads deep-reading-analyst automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

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
  2. "How deep do you want to go?"
    • Quick (15min) / Standard (30min) / Deep (60min) / Research (120min+)
  3. "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.

FAQ

What does the deep-reading-analyst skill do?

Comprehensive framework for deep analysis of articles, papers, and long-form content using 10+ thinking models (SCQA, 5W2H, critical thinking, inversion, mental models, first principles, systems thinking, six thinking hats). Use when users want to: (1) deeply understand complex articles/content, (2) analyze arguments and identify logical flaws, (3) extract actionable insights from reading materials, (4) create study notes or learning summaries, (5) compare multiple sources, (6) transform knowledge into practical applications, or (7) apply specific thinking frameworks. Triggered by phrases like 'analyze this article,' 'help me understand,' 'deep dive into,' 'extract insights from,' 'use [framework name],' or when users provide URLs/long-form content for analysis.

How do I install deep-reading-analyst?

Run npx skills add https://github.com/yyh211/claude-meta-skill --skill deep-reading-analyst in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/deep-reading-analyst/ (personal) or .claude/skills/deep-reading-analyst/ (project).

Which AI agents can use deep-reading-analyst?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is deep-reading-analyst safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains deep-reading-analyst?

It is published by yyh211 in yyh211/claude-meta-skill. It has 75 recorded installs.

More from yyh211/claude-meta-skill

View all →

daily-ai-news

yyh211/claude-meta-skill

Aggregates and summarizes the latest AI news from multiple sources including AI news websites and web search. Provides concise news briefs with direct links to original articles. Activates when user asks for 'today's AI news', 'AI updates', 'latest AI developments', or mentions wanting a 'daily AI briefing'.

frontend-design

yyh211/claude-meta-skill

Open the skill to read its SKILL.md instructions.

prompt-optimize

yyh211/claude-meta-skill

Expert prompt engineering skill that transforms Claude into "Alpha-Prompt" - a master prompt engineer who collaboratively crafts high-quality prompts through flexible dialogue. Activates when user asks to "optimize prompt", "improve system instruction", "enhance AI instruction", or mentions prompt engineering tasks.

AI & Agents72 installs

dry-refactoring

yyh211/claude-meta-skill

Guides systematic code refactoring following the DRY (Don't Repeat Yourself) principle. Use when user asks to eliminate code duplication, refactor repetitive code, apply DRY principle, or mentions code smells like copy-paste, magic numbers, or repeated logic. Implements a 4-step workflow from identifying repetition to verified refactoring.

create-skill-file

yyh211/claude-meta-skill

Guides Claude in creating well-structured SKILL.md files following best practices. Provides clear guidelines for naming, structure, and content organization to make skills easy to discover and execute.

fastgpt-workflow-generator

yyh211/claude-meta-skill

Open the skill to read its SKILL.md instructions.

Popular docs & writing skills

View category →

grill-with-docs

mattpocock/skills

A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.

Docs & Writing1.2M installs

lark-markdown

open.feishu.cn

飞书 Markdown:查看、创建、上传、编辑和比较飞书中的原生 Markdown 文件。当用户要操作飞书 Markdown 文件,或比较其远端版本及本地草稿时使用。纯本地 Markdown 文件操作不触发本 skill。不负责将 Markdown 导入为飞书在线文档,也不负责文件搜索、权限、评论、移动、删除等云空间管理操作。

Docs & Writing778.7K installs

writing-beats

mattpocock/skills

Writing, exploit; assemble raw material into a journey of beats, grounding each term before a beat leans on it.

Docs & Writing490.7K installs

writing-shape

mattpocock/skills

Writing, exploit: shape raw material into an article, paragraph by paragraph.

Docs & Writing490.2K installs

writing-fragments

mattpocock/skills

Writing, explore: mine raw fragments, no structure yet.

Docs & Writing489.6K installs

writing-for-agents

mattpocock/skills

Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.

Docs & Writing479.3K installs