explain

Generates diagram-rich markdown reports explaining how a scoped section of code works.

2|Updated Apr 4, 2026
One-click install
npx skills add https://github.com/DoctorMozg/claude-pipelines --skill explain-doctormozg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: explain
Source: https://github.com/DoctorMozg/claude-pipelines/tree/main/plugins/mz-research-pipe/skills/explain
Command: npx skills add https://github.com/DoctorMozg/claude-pipelines --skill explain-doctormozg

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Understanding unfamiliar or complex code requires hours of manual reading, and the resulting knowledge stays in one person's head. This Skill orchestrates a read-only deep-dive analysis that produces a comprehensive, diagram-rich report explaining how code works, why it is designed that way, and what potential issues exist. ## Core Features & Use Cases - Adaptive multi-agent research: Dispatches 1-3 parallel researcher agents (structure analyst, execution flow analyst, domain researcher) based on scope size and detected external dependencies. - Mandatory mermaid diagrams: Every report includes a high-level architecture diagram and a primary execution flow diagram, plus optional data flow, sequence, class, ER, state, and mindmap diagrams with strict quality checks. - Flexible scoping: Analyze by path, glob, git branch changes, working tree, or a free-text question like "how does the payment flow work". - Use Case: A developer joining a legacy project runs the skill on src/auth/ with the question "how does token refresh work" and receives a report with architecture diagrams, step-by-step execution traces with file:line references, design decision rationale, and observed potential issues. ## Quick Start Ask the assistant to explain how the WebSocket reconnection logic in src/auth/ works and write the report to the default output path.

Frequently Asked Questions about explain

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a diagram explaining how my code works?

Invoke the explain skill with a path or question, such as a directory like src/auth/ plus a focus question. It dispatches researcher agents to read the code and produces a markdown report with mandatory mermaid architecture and execution flow diagrams.

How to document an unfamiliar legacy codebase automatically?

Run the explain skill over the target directories or the whole repository with a focusing question. For scopes over 10 files it splits analysis between a structure analyst and a flow analyst, then compiles a report with component breakdowns, execution traces, and design decisions.

Can the explain skill modify or fix my code?

No, the explain pipeline is strictly read-only and never modifies source code. If you want code changed, debugged, or fixed, use the debug or build skills instead.

What diagram types does the code explanation report include?

Every report includes a high-level architecture flowchart and a primary execution flow diagram. Depending on the code, it also adds data flow, sequence, class, ER, state, and mindmap diagrams, each validated for mermaid syntax and node limits.

What happens when the explain scope is ambiguous or empty?

The skill asks you to clarify via a question prompt rather than guessing. If the resolved file list is empty it reports that and exits cleanly, and global scope always requires a focusing question.