AgentSkillsCN

mental-models

将查理·芒格的“思维模型拼图”应用于各类问题。当用户提出决策分析需求、说“帮我好好思考一下”、“运用某种思维模型”、提及具体模型名称（如逆向思维、瓶颈效应、二阶思维），或需要一套结构化的思维框架时，均可使用此方法。

SKILL.md
--- frontmatter
name: mental-models
description: Apply Charlie Munger's latticework of mental models to any problem. Use when user requests decision analysis, says "help me think", "apply mental model", mentions model names (inversion, bottlenecks, second-order thinking), or needs structured thinking frameworks.

Mental Models

Apply 98 cognitive frameworks from multiple disciplines to analyze problems, make decisions, and think more clearly.

Quick Reference

Models by Category:

  • •General Thinking (m01-m09): First principles, inversion, second-order thinking
  • •Science (m10-m29): Leverage, inertia, feedback loops, ecosystems
  • •Systems Thinking (m30-m40): Bottlenecks, scale, margin of safety, emergence
  • •Mathematics (m41-m47): Randomness, regression to mean, local vs global maxima
  • •Economics (m48-m59): Trade-offs, scarcity, creative destruction
  • •Art (m60-m70): Framing, audience, contrast
  • •Strategy (m71-m75): Asymmetric warfare, seeing the front
  • •Human Nature (m76-m98): Biases, incentives, social proof

How to Use

Quick Apply

When user mentions a specific model + context:

  1. •Load the model file from models/ directory
  2. •Read: Description, Thinking Steps, When to Avoid
  3. •Apply Thinking Steps to user's context
  4. •Provide 3-5 actionable insights
  5. •Note any cautions from "When to Avoid"

Example: "Apply inversion to this architecture decision"

Discovery

When user asks "what models for X":

  1. •Search resources/model-index.json for matching keywords
  2. •Return top 5 relevant models with brief descriptions
  3. •Offer to quick-apply any of them

Example: "What mental models help with scaling?"

Deep Analysis

When user has complex decision:

  1. •Ask 2-3 clarifying questions
  2. •Select max 3 most relevant models
  3. •Apply each model's Thinking Steps systematically
  4. •Synthesize insights across models
  5. •Provide actionable recommendations

Example: "Help me think through whether to accept this job offer"

Model File Structure

Each model in models/ contains:

  • •Description: Core concept
  • •When to Avoid: Limitations
  • •Keywords: Application contexts
  • •Thinking Steps: Sequential framework (follow exactly)
  • •Coaching Questions: Prompts for deeper exploration

Key Guidelines

  1. •Quality over quantity: Max 3 models per analysis
  2. •Follow Thinking Steps exactly: That's where the value is
  3. •Check "When to Avoid": Alert user if model doesn't fit
  4. •Latticework approach: Show how models connect
  5. •Be actionable: Provide concrete insights, not just theory

Resources

  • •resources/model-index.json - Searchable index of all 98 models
  • •resources/quick-reference.md - Common patterns and model combinations