Skip to content
Operations
Command

/docs-explanation

Create in-depth explanations that build conceptual understanding of complex topics

BOOST
From plugin
guidance-for-claude-code-with-amazon-bedrock
40923 skills25 agents23 commands
Install
> /plugin marketplace add aws-solutions-library-samples/guidance-for-claude-code-with-amazon-bedrock

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/docs-explanation

Context preview

What this command does when you run it.

Create in-depth explanations that build conceptual understanding of complex topics

Command definition

docs-explanation.md
name: docs-explanation
description: Create in-depth explanations that build conceptual understanding of complex topics
version: 1.0.0
argument-hint: "[concept to explain] [--architecture|--design|--theory|--comparison]"

Diataxis Explanation Command

You are in the **EXPLANATION** phase of the Diataxis documentation workflow. Your mission is to create understanding-oriented documentation that provides context, background, and deeper insights into concepts, designs, and decisions.

⚠️ **IMPORTANT**: This command is for creating EXPLANATORY documentation ONLY. Focus exclusively on:

  • Providing conceptual understanding and context
  • Explaining the "why" behind decisions
  • Discussing alternatives and trade-offs
  • Offering broader perspectives and connections
  • Documenting everything in `docs/explanation/[topic-slug].md`

Concept to Explain

$ARGUMENTS

If no specific concept was provided above, ask the user: "What concept, architecture, or design decision would you like explained in depth?"

🤔 Explanation Objectives

1. **Deepen Understanding**: Provide context and background 2. **Explain Rationale**: Why things are the way they are 3. **Discuss Alternatives**: What other approaches exist 4. **Make Connections**: How concepts relate to each other 5. **Provide Perspective**: Historical and future context

Extended Thinking Strategy

  • **Simple concepts**: Clear analogies and examples
  • **Complex systems**: Think about relationships and interactions
  • **Design decisions**: Think hard about trade-offs and constraints
  • **Architectural choices**: Ultrathink about long-term implications

Parallel Explanation Subagents

Deploy concurrent documentation specialists: @docs-explanation-agent @architecture-documenter @system-designer @business-analyst

All subagents work in parallel to create comprehensive explanations:

  • @docs-explanation-agent: Create conceptual, understanding-oriented content
  • @architecture-documenter: Document design decisions and rationale
  • @system-designer: Explain system architecture and patterns
  • @business-analyst: Provide business context and requirements background

Explanation Documentation Framework

Step 1: Define the Topic

## About [Topic]

### Overview
[High-level introduction to the concept]

### Why This Matters
- [Business impact]
- [Technical significance]
- [User benefits]

### Scope of This Explanation
This document explains:
- [Aspect 1]
- [Aspect 2]
- [Aspect 3]

This document does not cover:
- [Out of scope 1]
- [Out of scope 2]

Step 2: Provide Context

## Background and Context

### Historical Perspective
[How we got here - evolution of the concept]

### Current State
[Where we are now - current implementation/understanding]

### Industry Context
[How others approach this - standards and practices]

### Our Approach
[Why we chose this path - specific context]

Step 3: Core Concepts

## Understanding [Core Concept]

### The Fundamental Idea
[Explain the core concept in simple terms]

### Analogy
Think of [concept] like [familiar analogy]. Just as [analogy explanation], 
[concept] works by [parallel explanation].

### Key Principles
1. **Principle 1**: [Explanation]
   - Why it matters
   - How it works
   - Implications

2. **Principle 2**: [Explanation]
   - Why it matters
   - How it works
   - Implications

### Mental Model

[Visual or conceptual model] Component A → Process → Component B ↓ ↓ Storage Output

Step 4: Design Decisions

## Design Decisions and Trade-offs

### Decision: [Specific Choice Made]

#### Options Considered
1. **Option A** (Chosen)
   - Pros: [Benefits]
   - Cons: [Drawbacks]
   - Why chosen: [Reasoning]

2. **Option B**
   - Pros: [Benefits]
   - Cons: [Drawbacks]
   - Why not: [Reasoning]

3. **Option C**
   - Pros: [Benefits]
   - Cons: [Drawbacks]
   - Why not: [Reasoning]

#### Trade-offs Accepted
- We prioritized [quality] over [quality]
- We accepted [limitation] to gain [benefit]
- We chose [approach] knowing [consequence]

#### Future Considerations
- This decision allows for [future possibility]
- We may revisit if [condition changes]
- Migration path exists to [alternative]

Step 5: Relationships and Connections

## How This Relates to Other Concepts

### Relationship to [Related Concept 1]
[Explain connection and interaction]

### Relationship to [Related Concept 2]
[Explain connection and interaction]

### Part of Larger System

[Broader Context] ├── [This Concept] ├── [Related System] └── [Connected Component]


### Dependencies
- Depends on: [What this needs]
- Depended on by: [What needs this]
- Interfaces with: [What it connects to]

Explanation Deliverables

Output File Location

All explanation documentation will be generated in the `docs/explanation/` directory with descriptive filenames based on the concept being explained.

Explanation Template Structure

# Understanding [Topic]

## Introduction
[Accessible introduction that draws readers in]

## The Big Picture
[Context and significance in the broader landscape]

## Core Concepts

### What Is [Concept]?
[Clear, accessible explanation]

### Why [Concept] Exists
[Problem it solves, need it addresses]

### How [Concept] Works
[Conceptual overview, not implementation details]

## Design Philosophy

### Guiding Principles
[What drives the design]

### Architectural Decisions
[Key choices and their rationale]

### Trade-offs
[What we optimized for vs. what we sacrificed]

## Alternatives and Comparisons

### Alternative Approaches
[Other ways to solve the same problem]

### When to Use Which
[Decision framework for choosing approaches]

### Evolution and History
[How approaches have evolved]

## Practical Implications

### Impact on Development
[How this affects day-to-day work]

### Impact on Users
[How this affects end users]

### Impact on Operat
Read more
Ships withguidance-for-claude-code-with-amazon-bedrock

This guidance enables enterprise deployment of Claude Code and Claude Cowork on Amazon Bedrock across command-line (CLI) and desktop surfaces — with secure single sign-on (SSO), usage monitoring, and cost controls.

Get the whole plugin