Multi-Perspective In-Depth Analysis

Use 10 independent Sub-agents to act as "consultants" representing the world's top minds, conduct parallel analysis on the material provided by the user, then cross-summarize to produce a structured report.

Lineup: Dan Sullivan, Elon Musk, Zhang Xiaolong, MrBeast, Charlie Munger, Peter Thiel, Steve Jobs, Jeff Bezos, Zhang Yiming, Ren Zhengfei

Core Value: High credibility when independent analyses converge on conclusions; differences in conclusions often represent the most valuable insights.

Process Overview

Phase	Name	Objective	Execution Method
1	Requirement Convergence	Clarify what to analyze, why to analyze, and which perspectives to use	Dialogue
2	Material Preparation	Organize analysis materials and confirm input for each agent	Dialogue confirmation
3	Parallel Analysis	10 Sub-agents conduct independent analysis and produce individual reports	Agent Tool Parallel Execution
4	Cross-Summary	Identify consensus, differences, and extract actionable recommendations	Main dialogue
5	Report Output	Structured report + Optional: Save as file	Main dialogue

Phase 1: Requirement Convergence

Objective: Clarify three things — what to analyze, why, and which perspectives to use.

Information That Must Be Confirmed

1. Analysis Object

What does the user want to analyze? (Review data over a period, a product strategy, a business direction, a decision, etc.)
Are there any existing material files? (If yes, ask the user to provide the path or content)

2. Analysis Purpose

What does the user want to gain from the analysis? (Diagnose problems, verify directions, discover blind spots, make decisions, etc.)
Are there any core questions the user specifically wants answered?

3. Perspective Selection

10 built-in perspectives are provided by default (see "Built-in Perspective Library" below). Users can:

Use all 10 default perspectives: Suitable for comprehensive diagnosis ("The world's 10 smartest minds act as your consultants simultaneously")
Select 3-5 of them: Suitable for focusing on specific dimensions
Customize perspectives: Users specify other thinkers/frameworks, and AI temporarily builds corpora via WebSearch

If the user is unsure what to choose, recommend based on the analysis object:

Analysis Object Type	Recommended Perspectives
Personal/Creator Comprehensive Review	All 10 perspectives
Product Strategy	Elon Musk (First Principles) + Zhang Xiaolong (Product Intuition) + Steve Jobs (Ultimate Experience) + Jeff Bezos (Customer Obsession)
Content Growth	MrBeast (Growth) + Zhang Xiaolong (Authenticity of Demand) + Zhang Yiming (Algorithmic Thinking) + Ren Zhengfei (Pressure Principle)
Business Decision	Charlie Munger (Mental Models) + Peter Thiel (Monopoly Thinking) + Jeff Bezos (Flywheel) + Dan Sullivan (10x Thinking)
Career Direction	Dan Sullivan (10x vs 2x) + Elon Musk (Anti-consensus) + Peter Thiel (Secrets) + Ren Zhengfei (Long-term Hard Work)
Competitive Strategy	Ren Zhengfei (Pressure Principle + Pinpoint Strategy) + Peter Thiel (Monopoly) + Charlie Munger (Circle of Competence)
Strategic Review	Ren Zhengfei (Huawei's Winter + Self-Criticism) + Charlie Munger (Think in Reverse) + Elon Musk (First Principles)

Criteria for Completing Convergence

Clear analysis object (with materials or clear description)
Clear analysis purpose (at least one core question)
Confirmed perspectives

Phase 2: Material Preparation

Objective: Organize the materials provided by the user into inputs that each agent can directly use.

Material Organization Rules

If the user provides a file path: Read the file content and organize it into a text summary
If the user describes in dialogue: Extract key information to form a structured "Analysis Brief"
If the material is too long (over 5000 words): Extract core data and key facts, compress into a version digestible for each agent

Input Package for Each Agent

Each Sub-agent must receive input including:

Role Setting: You are [Thinker's Name], analyze using the methodology of [Framework Name]
Analysis Material: [Organized user material]
Core Question: The user's most desired question to answer
Framework Reference: Full original text of the corresponding reference file
Output Requirements:

⚠️ Hard Rules for Framework Reference:

Do not pre-inject the full reference text into the main prompt — stuffing 5 corpora into the prompt will make it too long, causing agent loss or degradation
Instead: In each agent's prompt, tell it to read the corresponding reference file on its own
Each agent only reads its own file, keeping the workload manageable
The reference file includes original quotes from the thinker, specific cases, and analysis question templates — the agent must conduct analysis based on these
- Must use core concepts and terminology of the framework
- Must provide clear judgments and suggestions, avoid ambiguity
- Must include "The Harshest Truth" — a diagnosis the user may not want to hear but must listen to
- No word limit; write as much as needed
- Tone: Speak directly to the user in the first person of the thinker, like a real consultant

Confirm and Enter Parallel Analysis

Phase 3: Parallel Analysis (Core Phase)

Execution Method: Use Agent tools to start all Sub-agents in parallel at once.

Pre-launch Preparation: Create Output Folder First

Before launching the agents, must create the output folder first:

bash

mkdir -p "多视角分析/分析报告-[主题简称]"

Only after the folder is created can the agents be launched. Each agent's prompt will instruct it to write the report to this folder.

Sub-agent Launch Template

⚠️ Core Strategy: Each agent reads its own corpus and writes its own report

The main dialogue does not pre-read reference files (to avoid prompt length causing agent loss)
Each agent reads its own reference file independently
After analysis, each agent writes its report to the specified file on its own
The main agent only reads all reports at the end for cross-summary

Corpus files and output files corresponding to each perspective:

#	Perspective	Reference File	Output File
1	Dan Sullivan	`.claude/skills/multi-perspective-analysis/reference/01-10x-growth-dan-sullivan.md`	`多视角分析/分析报告-[主题]/01-DanSullivan.md`
2	Elon Musk	`.claude/skills/multi-perspective-analysis/reference/02-first-principles-elon-musk.md`	`多视角分析/分析报告-[主题]/02-马斯克.md`
3	Zhang Xiaolong	`.claude/skills/multi-perspective-analysis/reference/03-product-intuition-zhang-xiaolong.md`	`多视角分析/分析报告-[主题]/03-张小龙.md`
4	MrBeast	`.claude/skills/multi-perspective-analysis/reference/04-creator-growth-mrbeast.md`	`多视角分析/分析报告-[主题]/04-MrBeast.md`
5	Charlie Munger	`.claude/skills/multi-perspective-analysis/reference/05-mental-models-charlie-munger.md`	`多视角分析/分析报告-[主题]/05-芒格.md`
6	Peter Thiel	`.claude/skills/multi-perspective-analysis/reference/06-zero-to-one-peter-thiel.md`	`多视角分析/分析报告-[主题]/06-蒂尔.md`
7	Steve Jobs	`.claude/skills/multi-perspective-analysis/reference/07-product-vision-steve-jobs.md`	`多视角分析/分析报告-[主题]/07-乔布斯.md`
8	Jeff Bezos	`.claude/skills/multi-perspective-analysis/reference/08-customer-obsession-jeff-bezos.md`	`多视角分析/分析报告-[主题]/08-贝佐斯.md`
9	Zhang Yiming	`.claude/skills/multi-perspective-analysis/reference/09-algorithm-thinking-zhang-yiming.md`	`多视角分析/分析报告-[主题]/09-张一鸣.md`
10	Ren Zhengfei	`.claude/skills/multi-perspective-analysis/reference/10-strategic-survival-ren-zhengfei.md`	`多视角分析/分析报告-[主题]/10-任正非.md`

Prompt template for each agent:

You are [Thinker's Name]. Use your [Framework Name] to analyze the following material and answer the user's core question.

## Step 1: Read Your Analysis Framework (Mandatory)

First, use the Read tool to read the following file, which contains your complete analysis framework corpus:
[Absolute path to the corresponding reference file]

After reading, strictly base your analysis on the framework, original quotes, cases, and analysis questions in this file. Do not skip this step, and do not replace the file content with your own knowledge.

## Analysis Material

[Insert organized user material]

## User's Core Question

[The user's most desired question to answer]

## Output Requirements
1. Conduct in-depth analysis using the core concepts of your framework; avoid superficial discussions
2. Must quote original content from the reference file to support your analysis (marked with > quote format)
3. Provide clear judgments and suggestions
4. Speak directly to the user in the first person, like a real consultant
5. Must include a section of "The Harshest Truth" — a diagnosis that the user most needs to hear but may least want to listen to
6. Write as much as needed; prioritize depth over length control
7. Please output in Chinese

## Final Step: Save the Report (Mandatory)

After completing the analysis, use the Write tool to save your complete analysis report to the following file:
[Absolute path to the corresponding output file]

Do not skip this step. Your task is completed after writing.

Key Rules

Create folder first before launching: Only start agents after mkdir is completed
Must launch all in parallel: All selected agents must be launched in the same message. If 5 perspectives are selected, 5 agents must be launched without omission
Each agent writes its own file: The main agent is not responsible for writing perspective-specific reports, only the summary
Each agent is completely independent: Do not let any agent see the output of other agents
Do not preset conclusions: Do not imply "You should find XX problem" in the prompt

Phase 4: Cross-Summary

Objective: Extract high-value insights from the saved report files of each perspective.

⚠️ Execution Prerequisite: After all agents are completed, use the Read tool to read each report file in the output folder one by one to confirm they have all been written. Then conduct cross-analysis based on the file content.

Summary Framework

1. Consensus Findings (Agreement among multiple perspectives = High credibility)

Scan all reports to find points where 3 or more perspectives reach the same conclusion
These are the most reliable diagnoses — convergence of independent analyses indicates "hard facts"

2. Interesting Differences (Perspective conflicts = High-value insights)

Identify points where different perspectives provide contradictory suggestions
Analyze the reasons for differences: Are they due to different framework assumptions? Or different levels of the problem being observed?
Differences are often more enlightening than consensus — they expose the multi-faceted nature of the problem

3. Exclusive Insights (Proposed by only one perspective)

Did any perspective propose an important point that no one else mentioned?
These may be overlooked blind spots

4. Actionable Recommendations (Converge into actions)

Extract specific action recommendations from all reports
Sort by "Importance x Executability"
Converge into 3-5 most critical next steps

Summary Output Format

markdown

# [Analysis Topic] — Multi-Perspective Diagnostic Report

> Analysis Date: YYYY-MM-DD
> Perspectives: [Perspectives used]
> Material: [What was analyzed]

---

## Multi-Person Consensus (High Credibility)
[3-5 core conclusions agreed by most perspectives]

## Interesting Differences
[Conflicting points between perspectives and analysis of their causes]

## Exclusive Insights
[Important points proposed only by a specific perspective]

## The Harshest Truth (One per perspective)
| Perspective | The Harshest Truth |
|------|------------|
| ... | ... |

## Next Steps (3-5 items)
[Specific action recommendations sorted by priority]

---

## Appendix: Full Reports from Each Perspective
[Keep the original output of each agent, users can expand to read]

Phase 5: Saving and Output (Automatic Execution, No User Request Needed)

⚠️ This is a mandatory step; it must be executed automatically after agents finish, do not wait for user requests.

Saving Process

Step 1: Create Output Folder

Create a folder named after the analysis topic under the

多视角分析/

directory:

多视角分析/
  分析报告-[主题简称]/
    00-汇总报告.md
    01-Dan-Sullivan.md
    02-Elon-Musk.md
    03-Naval-Ravikant.md
    04-张小龙.md
    05-MrBeast.md

Step 2: Confirm All Perspective Reports Have Been Written

Each agent has already written its report to the folder (completed in Phase 3). Use ls or Glob to confirm all files exist. If any agent fails to write, the main agent will补写 from the agent's returned content.

Step 3: Main Agent Writes Summary Report

Based on the cross-analysis in Phase 4, write the summary report to

00-汇总报告.md

. The main agent only writes this one file.

Step 4: Inform the User

After saving, tell the user:

Which directory the files are saved in
Total number of reports
Display the core content of the summary report in the dialogue (consensus + differences + action recommendations)

Naming Rules

Folder name:

分析报告-[主题简称]

(e.g.,

分析报告-职业决策

分析报告-4月策略

)

If a folder with the same topic already exists, add the date after the topic:
```
分析报告-职业决策-0410
```
If there are custom perspectives (corpora built via WebSearch), save them to the reference directory for reuse

Built-in Perspective Library (10 World Top Minds)

Perspective 1: Dan Sullivan — 10x Growth

Reference File:

reference/01-10x-growth-dan-sullivan.md

Core Dimensions: 10x vs 2x Filter, 20%/80% Rule, 4C Cycle (Commitment→Courage→Capability→Confidence), Unique Ability, Want vs Need Best For: Personal growth strategy, time allocation, direction selection, career decisions

Perspective 2: Elon Musk — First Principles

Reference File:

reference/02-first-principles-elon-musk.md

Core Dimensions: First Principles Hypothesis Breaking, Five-Step Engineering Algorithm, Probability x Importance, Anti-consensus Judgment, Time Allocation Best For: Product strategy, process optimization, resource allocation decisions

Perspective 3: Zhang Xiaolong — Product Intuition

Reference File:

reference/03-product-intuition-zhang-xiaolong.md

Core Dimensions: Authenticity of Demand, Restraint in Action, Use and Leave, Forest vs Palace, KPI as By-product Best For: Product design, content strategy, system evaluation

Perspective 4: MrBeast — Creator Growth

Reference File:

reference/04-creator-growth-mrbeast.md

Core Dimensions: Title/Click-through Rate, Topic Breadth Testing, Content Flywheel, Output and Iteration, Independent Product Thinking Best For: Content growth, topic evaluation, dissemination strategy

Perspective 5: Charlie Munger — Mental Models

Reference File:

reference/05-mental-models-charlie-munger.md

Core Dimensions: Interdisciplinary Mental Models, Think in Reverse, Circle of Competence, Misjudgment Psychology, Lollapalooza Effect, Checklist, Wait for Fat Pitch, Avoid Stupidity Best For: Decision quality diagnosis, cognitive bias audit, long-term strategy

Perspective 6: Peter Thiel — Zero to One / Monopoly Thinking

Reference File:

reference/06-zero-to-one-peter-thiel.md

Core Dimensions: Zero to One vs One to N, Monopoly Thinking, Reverse Thinking/Secrets, Power Law, Definite Optimism, Last Mover Advantage, Importance of Sales Best For: Positioning and differentiation, competitive strategy, direction selection

Perspective 7: Steve Jobs — Ultimate Experience and Narrative

Reference File:

reference/07-product-vision-steve-jobs.md

Core Dimensions: Ultimate Simplicity, Product as Marketing, Reality Distortion Field, Connecting the Dots, Taste and Aesthetics, Focus on Saying No, Think Different Best For: Product experience, brand narrative, content quality

Perspective 8: Jeff Bezos — Customer Obsession and Flywheel

Reference File:

reference/08-customer-obsession-jeff-bezos.md

Core Dimensions: Day 1 Thinking, Customer Obsession/Reverse Working Method, Flywheel Effect, Long-termism, High Standards/Six-Page Memo, One-way Door vs Two-way Door, Regret Minimization Best For: Growth strategy, user insights, long-term planning

Perspective 9: Zhang Yiming — Algorithmic Thinking

Reference File:

reference/09-algorithm-thinking-zhang-yiming.md

Core Dimensions: Delayed Gratification, Algorithm/Information Distribution Efficiency, Think Like a Machine, Always Entrepreneurship, Context not Control, Talent Density, Global Vision Best For: Personal growth, organizational efficiency, information distribution strategy

Perspective 10: Ren Zhengfei — Strategic Survival Philosophy

Reference File:

reference/10-strategic-survival-ren-zhengfei.md

Core Dimensions: Pressure Principle (Focus All Efforts on One Point), Huawei's Winter (Crisis Awareness), Striver-oriented, Gray Management, Deepen the Beach and Lower the Weir, Pinpoint Strategy (Focus on Main Channel), Self-Criticism Best For: Strategic focus, crisis management, resource allocation, competitive strategy, long-term survival

Custom Perspectives

If users want to add perspectives outside the built-in library, follow the process in

reference/_how-to-create-reference.md

. This document includes corpus file templates, collection strategies, quality standards, and checklists.

Core Principles

Independence is the lifeline: Each agent must conduct completely independent analysis and cannot see the output of other agents. This is the entire value of multi-perspective analysis
Must be parallel: All agents must be launched in the same message, not serially. Serial execution will cause later agents to be influenced by earlier ones
Depth first: It is better to have a 3000-word in-depth report for each perspective than 5 shallow 500-word summaries
Must have sharpness: Each agent must provide "The Harshest Truth". Polite but useless analysis is better not done
Summary must have opinions: Cross-summary is not simple listing; analyze why consensus formed and why differences occurred
User decision-making power: Analysis is a tool, decision-making belongs to the user. Provide suggestions at the end of the report, do not make decisions

multi-perspective-analysis

NPX Install

Tags

SKILL.md Content (Chinese)

Multi-Perspective In-Depth Analysis

Process Overview

Phase 1: Requirement Convergence

Information That Must Be Confirmed

Criteria for Completing Convergence

Phase 2: Material Preparation

Material Organization Rules

Input Package for Each Agent

Confirm and Enter Parallel Analysis

Phase 3: Parallel Analysis (Core Phase)

Pre-launch Preparation: Create Output Folder First

Sub-agent Launch Template

Key Rules

Phase 4: Cross-Summary

Summary Framework

Summary Output Format

Phase 5: Saving and Output (Automatic Execution, No User Request Needed)

Saving Process

Naming Rules

Built-in Perspective Library (10 World Top Minds)

Perspective 1: Dan Sullivan — 10x Growth

Perspective 2: Elon Musk — First Principles

Perspective 3: Zhang Xiaolong — Product Intuition

Perspective 4: MrBeast — Creator Growth

Perspective 5: Charlie Munger — Mental Models

Perspective 6: Peter Thiel — Zero to One / Monopoly Thinking

Perspective 7: Steve Jobs — Ultimate Experience and Narrative

Perspective 8: Jeff Bezos — Customer Obsession and Flywheel

Perspective 9: Zhang Yiming — Algorithmic Thinking

Perspective 10: Ren Zhengfei — Strategic Survival Philosophy

Custom Perspectives

Core Principles