Agent skill
decomposing-complex-problems
Deconstructs complex problems using the 'Polymath Investor' framework with a 5-step cognitive chain (First Principles → Modular Isolation → Pareto Filtering → Cross-Domain Mapping → Dynamic Zooming). Use when analyzing systems, architectures, businesses, or learning challenges; when seeking high-leverage insights or 'Alpha'; when applying first principles thinking; or when needing cross-disciplinary perspectives to reveal blind spots. Transforms overwhelming complexity into actionable levers.
Install this agent skill to your Project
npx add-skill https://github.com/bamecho/my-claude-skills/tree/main/skills/decomposing-complex-problems
SKILL.md
Polymath Decomposition Protocol
Overview
A high-fidelity cognitive tool for breaking down complexity. It rejects linear "specialist" thinking in favor of a "lattice of mental models."
Goal: Transform "Overwhelming Complexity" into "Actionable Levers."
Critical Rules
- NO ANALOGY (Initially) - Phase 1 must be pure fact/physics. Analogies are forbidden until Phase 4.
- RUTHLESS REDUCTION - If it doesn't drive the outcome, it is noise. Ignore it.
- FORCED ISOMORPHISM - Find a structural match in a non-related field (Biology/Physics/History).
The 5-Phase Process
PHASE 1: THE LEGO DEMOLITION (First Principles)
"Decompose to the basic truths."
- Action: Smash the problem/system into its smallest indivisible units ("The LEGO Bricks").
- Constraint: Ignore "Best Practices" or "Industry Standards".
- Question: What are the physical, mathematical, or economic limits that cannot be violated?
- Output: A list of 3-5 "Atomic Facts" (e.g., "Data travels at speed of light," "Users have finite attention").
PHASE 2: ISOLATION & DECOUPLING (Divide & Conquer)
"Make it digestible."
- Action: Group atomic units into independent "Black Boxes" (Modules).
- Constraint: Modules must be Decoupled. Analyzing Module A should not require holding Module B in memory.
- Question: What are the clear "Interfaces" between these boxes? Where do they touch?
- Output: A defined list of Sub-systems and their Inputs/Outputs.
PHASE 3: THE PARETO FILTER (Key Drivers)
"Ignore the noise."
- Action: Identify the "High Convexity" modules.
- Constraint: 80% of the modules are "Maintenance/Noise". Ignore them. Focus on the 20% that generate value or risk.
- Question: If this module is deleted, does the system die? If this module improves by 1%, does the outcome improve by 10%?
- Output: Identification of the single "Bottleneck" or "Growth Engine."
PHASE 4: STRUCTURAL MAPPING (The Polymath Lens)
"Find the isomorphism."
- Action: Force a metaphor from a completely alien domain onto the "Key Driver" identified in Phase 3.
- Domains to Scan:
- Physics/Thermodynamics: (Entropy, Friction, Leverage, Potential Energy)
- Evolutionary Biology: (Adaptation, Parasitism, Symbiosis, Immune Response)
- History/Warfare: (Empires, Supply Lines, Guerilla Tactics)
- Question: "What is this system's equivalent in Nature?" (e.g., "This microservice architecture looks like a hive mind," or "This marketing funnel has high 'kinetic friction'.")
- Output: A "Cross-Domain Insight" that reveals a blind spot.
PHASE 5: DYNAMIC ZOOM (Zoom In & Out)
"See the forest and the bark."
- Action: Oscillate between the texture and the landscape.
- Zoom In: Look at the "Code/Texture" of the Key Driver. (The implementation detail).
- Zoom Out: Look at the "Time/Cycle" of the system. (Where is this in the historical lifecycle?).
- Output: A synthesis of how the micro-detail affects the macro-destiny.
Output Format
Present the analysis using this structure:
### 1. 乐高拆解 (First Principles)
* **原子事实**: [不可再分的物理/逻辑真相]
* **原子事实**: ...
### 2. 模块隔离 (Decoupling)
* [模块A] -> (接口) -> [模块B]
* **解耦说明**: [为何这两个模块可以独立看待]
### 3. 帕累托杠杆 (The 20%)
* **关键驱动因子**: [系统生死的决定性模块]
* **被忽略的噪音**: [可以暂时无视的80%细节]
### 4. 跨学科映射 (Structural Mapping)
* **映射领域**: [例如:生物免疫系统]
* **同构性分析**: [系统的Bug处理机制就像...因此它缺少...]
* **盲点揭示**: [基于这个比喻,我们要警惕...]
### 5. 变焦综合 (Zoom In/Out)
* **微观纹理**: [代码/执行层面的具体观察]
* **宏观定位**: [该系统处于历史周期的哪个阶段]
* **下一步行动**: [基于上述所有分析的一个高杠杆动作]
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
decomposing-complex-problems
Deconstructs complex problems using the 'Polymath Investor' framework with a 5-step cognitive chain (First Principles → Modular Isolation → Pareto Filtering → Cross-Domain Mapping → Dynamic Zooming). Use when analyzing systems, architectures, businesses, or learning challenges; when seeking high-leverage insights or 'Alpha'; when applying first principles thinking; or when needing cross-disciplinary perspectives to reveal blind spots. Transforms overwhelming complexity into actionable levers.
deep-analysis
Execute high-density architectural analysis on user ideas. Move from 'Vague' to 'Verified' using a 5-step logic chain: Calibration → Decomposition → Excavation → Re-Architecting → Inversion. This skill should be used when analyzing system architecture, validating technical ideas, or performing pre-mortems on solutions.
nexus-prompt-engineer
4-D prompt engineering assistant that transforms vague requirements into high-precision prompts through guided interaction. Trigger when users need to: (1) craft high-quality system prompts, (2) optimize existing prompts, (3) use '/fast' for quick generation or '/audit' for prompt review. Applicable to any scenario requiring carefully designed prompts.
round-table-analysis
Deep pre-execution analysis framework using four philosophers (Socrates, Aristotle, Feynman, Drucker) for structured critical thinking before implementing plans. Use when users need to thoroughly understand a topic, question assumptions, clarify objectives, or explore multiple perspectives before taking action. Triggered by phrases like "help me think through," "analyze this topic," "before we start," "I want to understand," or when complex decisions require deeper exploration.
brainstorming
Use this when the user explicitly asks to brainstorm, or when the task is clearly about upfront design, solution exploration, or requirement shaping before implementation. Avoid loading it for routine implementation work unless that exploratory step is obviously needed.
planning-with-files
Implements Manus-style file-based planning for complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when starting complex multi-step tasks, research projects, or any task requiring >5 tool calls. Now with automatic session recovery after /clear.
Didn't find tool you were looking for?