返回 Skill 列表
extension
分类: AI Agent 能力无需 API Key

domain-expert

快速学习新领域的元技能。涵盖领域研究过程、专家咨询模拟、知识验证和文档生成。

person作者: jakexiaohubgithub

Domain Expert

Overview

A meta-skill for systematically learning and applying expertise in unfamiliar domains. Enables rapid acquisition of domain knowledge through structured research, expert simulation, and knowledge validation.

When to Use

  • Entering an unfamiliar technical domain
  • Building products in new industries
  • Understanding complex regulations or standards
  • Bridging knowledge gaps before implementation
  • Validating assumptions about a domain

Quick Reference

| Phase | Duration | Output | |-------|----------|--------| | Survey | 1-2 hours | Domain landscape map | | Deep Dive | 2-4 hours | Core concepts mastery | | Expert Simulation | 1-2 hours | Validated understanding | | Documentation | 1 hour | Reusable knowledge base |


Phase 1: Domain Survey

Initial Landscape Mapping

## Domain Survey: [Domain Name]

### What is this domain?
[One paragraph explaining the domain to a complete outsider]

### Who operates in this domain?
- **Practitioners**: [Who does the work]
- **Consumers**: [Who benefits]
- **Regulators**: [Who sets rules]
- **Vendors**: [Who sells tools/services]

### Core Problems Solved
1. [Problem 1]
2. [Problem 2]
3. [Problem 3]

### Key Terminology
| Term | Definition | Example |
|------|------------|---------|
| [Term 1] | [Definition] | [Usage] |
| [Term 2] | [Definition] | [Usage] |

### Related Domains
- [Adjacent domain 1] - [Relationship]
- [Adjacent domain 2] - [Relationship]

Research Sources Hierarchy

| Priority | Source Type | Purpose | |----------|-------------|---------| | 1 | Official Documentation | Authoritative definitions | | 2 | Industry Standards | Best practices | | 3 | Academic Papers | Deep understanding | | 4 | Expert Blogs | Practical insights | | 5 | Community Discussions | Real-world problems | | 6 | Books | Comprehensive coverage |

Questions to Answer

## Domain Understanding Checklist

### Fundamentals
- [ ] What problem does this domain solve?
- [ ] What are the core concepts/primitives?
- [ ] What are the key relationships between concepts?
- [ ] What are common misconceptions?

### Practical
- [ ] What tools do practitioners use?
- [ ] What are typical workflows?
- [ ] What are common failure modes?
- [ ] What metrics define success?

### Context
- [ ] What drives change in this domain?
- [ ] What are current trends?
- [ ] What regulations apply?
- [ ] Who are thought leaders?

Phase 2: Deep Dive

Concept Decomposition

## Core Concept: [Concept Name]

### Definition
[Precise definition in simple terms]

### Why It Matters
[Practical importance]

### Components
1. [Sub-concept 1]: [Explanation]
2. [Sub-concept 2]: [Explanation]
3. [Sub-concept 3]: [Explanation]

### Relationships
- Depends on: [Other concepts]
- Enables: [Other concepts]
- Conflicts with: [Other concepts]

### Examples
**Good Example**: [Description]
**Bad Example**: [Anti-pattern]

### Common Mistakes
1. [Mistake]: [Why it's wrong]
2. [Mistake]: [Why it's wrong]

Knowledge Mapping

                    [Domain]
                        |
        ________________|________________
        |               |               |
    [Concept A]    [Concept B]    [Concept C]
        |               |               |
    ___|___         ___|___         ___|___
    |     |         |     |         |     |
  [A.1] [A.2]     [B.1] [B.2]     [C.1] [C.2]

Depth Calibration

| Level | Knowledge | Application | |-------|-----------|-------------| | L1: Aware | Know it exists | Can recognize | | L2: Familiar | Understand basics | Can discuss | | L3: Competent | Know how to use | Can implement | | L4: Proficient | Know when/why | Can optimize | | L5: Expert | Deep understanding | Can innovate |

## Target Knowledge Levels

For [Project/Goal]:
- [Concept A]: L3 (need to implement)
- [Concept B]: L2 (need to integrate)
- [Concept C]: L1 (just awareness)

Phase 3: Expert Simulation

The Socratic Method

## Expert Interview Simulation

### Question 1: [Basic Question]
**Expected Expert Answer**: [How an expert would respond]
**My Understanding**: [What I think the answer is]
**Gap**: [What I'm missing]

### Question 2: [Edge Case]
**Expected Expert Answer**: [How an expert would respond]
**My Understanding**: [What I think the answer is]
**Gap**: [What I'm missing]

### Question 3: [Why Question]
**Expected Expert Answer**: [How an expert would respond]
**My Understanding**: [What I think the answer is]
**Gap**: [What I'm missing]

Devil's Advocate Analysis

## Challenging Assumptions

### Assumption 1: [Statement I believe]
**Strongest Counterargument**: [Best case against it]
**Response**: [How to address it]
**Verdict**: [Keep/Modify/Discard]

### Assumption 2: [Statement I believe]
**Strongest Counterargument**: [Best case against it]
**Response**: [How to address it]
**Verdict**: [Keep/Modify/Discard]

Expert Persona Templates

## Expert Persona: [Domain] Specialist

### Background
- 10+ years in [domain]
- Worked at [typical companies/contexts]
- Expertise in [specific areas]

### Common Advice This Expert Gives
1. [Typical guidance 1]
2. [Typical guidance 2]
3. [Typical guidance 3]

### Red Flags This Expert Watches For
1. [Warning sign 1]
2. [Warning sign 2]
3. [Warning sign 3]

### Questions This Expert Would Ask
1. [Diagnostic question 1]
2. [Diagnostic question 2]
3. [Diagnostic question 3]

Validation Techniques

| Technique | Purpose | Method | |-----------|---------|--------| | Teach-back | Test understanding | Explain to someone new | | Application | Test practical use | Build something small | | Comparison | Test relationships | Compare to known domain | | Edge Cases | Test depth | Find where knowledge breaks | | Prediction | Test mental model | Predict outcomes |


Phase 4: Knowledge Documentation

Domain Knowledge Base Template

# [Domain] Knowledge Base

## Overview
[2-3 paragraphs explaining the domain comprehensively]

## Core Concepts

### [Concept 1]
[Detailed explanation with examples]

### [Concept 2]
[Detailed explanation with examples]

## Key Terminology
[Glossary of important terms]

## Common Patterns
[Recurring solutions and approaches]

## Anti-patterns
[Common mistakes to avoid]

## Tools & Resources
[List of useful tools, docs, communities]

## Decision Framework
[How to make common decisions in this domain]

## FAQ
[Frequently asked questions with answers]

## Further Reading
[Books, articles, courses for deeper learning]

Cheat Sheet Format

# [Domain] Cheat Sheet

## Quick Reference

| Situation | Approach |
|-----------|----------|
| [When X] | [Do Y] |
| [When A] | [Do B] |

## Key Formulas/Patterns


## Common Mistakes

1. [Mistake] -> [Correct approach]
2. [Mistake] -> [Correct approach]

## Gotchas

- [Non-obvious thing 1]
- [Non-obvious thing 2]

Research Strategies

Breadth-First Research

1. Read 5-10 overview articles
2. Identify recurring themes
3. Map key concepts
4. Note terminology
5. List questions

Depth-First Research

1. Pick ONE core concept
2. Read primary sources
3. Work through examples
4. Test understanding
5. Move to next concept

Just-In-Time Learning

1. Start with specific goal
2. Research only what blocks you
3. Implement immediately
4. Learn from errors
5. Deepen as needed

Comparative Learning

## [New Domain] vs [Known Domain]

| Aspect | [Known] | [New] |
|--------|---------|-------|
| Core Problem | [X] | [Y] |
| Main Actors | [X] | [Y] |
| Key Trade-offs | [X] | [Y] |
| Success Metrics | [X] | [Y] |

### Key Differences
1. [Difference]
2. [Difference]

### Transferable Knowledge
1. [What carries over]
2. [What carries over]

Quality Indicators

Signs of Good Understanding

- [ ] Can explain to a beginner
- [ ] Can predict outcomes
- [ ] Can identify mistakes
- [ ] Can propose improvements
- [ ] Can connect to other domains
- [ ] Can answer "why" questions
- [ ] Can recognize edge cases

Signs of Shallow Understanding

- Only memorized facts
- Can't explain reasoning
- Confused by variations
- No mental model for predictions
- Can't identify bad advice

Knowledge Confidence Scale

| Level | Indicator | |-------|-----------| | High | Can teach and debate | | Medium | Can apply and explain | | Low | Can recognize and follow | | None | Need to look everything up |


Common Patterns

The Feynman Technique

1. Choose a concept
2. Explain it simply (as if to a child)
3. Identify gaps in explanation
4. Research to fill gaps
5. Simplify and refine
6. Repeat until clear

Mental Model Building

## Mental Model: [Domain]

### Core Metaphor
[Domain] is like [familiar thing] because [similarity]

### Key Dynamics
- When [X happens], [Y results]
- When [A increases], [B decreases]

### Boundaries
- This model breaks when [edge case]
- Does not explain [limitation]

Knowledge Transfer Template

## Transferring [Domain] Knowledge

### For Developers
[Technical framing of domain concepts]

### For Business Stakeholders
[Business framing of domain concepts]

### For Users
[User-friendly framing of domain concepts]

Red Flags - STOP

Never:

  • Assume expertise from surface reading
  • Skip primary sources for summaries only
  • Ignore domain experts' warnings
  • Apply patterns without understanding context
  • Trust outdated sources for evolving domains

Always:

  • Validate understanding through application
  • Seek multiple perspectives
  • Document gaps in knowledge
  • Update knowledge base as you learn
  • Cite sources for important claims

Integration

Related skills: brainstorm, plan, session Process: Use with plan for domain-specific implementations