explain

Explains unfamiliar code with layered, six-section breakdowns of specified files, directories, symbols, or concepts.

9|2|Updated Mar 7, 2026
One-click install
npx skills add https://github.com/harnessprotocol/harness-kit --skill explain-harnessprotocol
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: explain
Source: https://github.com/harnessprotocol/harness-kit/tree/main/plugins/explain/skills/explain
Command: npx skills add https://github.com/harnessprotocol/harness-kit --skill explain-harnessprotocol

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you quickly understand unfamiliar code by turning a file, directory, symbol, or concept into a structured explanation that clarifies purpose, relationships, and where to safely make changes.

Core Features & Use Cases

  • Layered, six-section explanations that cover summary, key components, connections, patterns and conventions, gotchas, and entry points for change.
  • Adaptive depth that chooses the right level of detail based on what you’re pointing at (single function vs file vs directory vs concept).
  • Project-aware grounding by using the repository’s CLAUDE.md Architecture section when available to match team terminology and conventions.
  • Map-first for wider scopes for directories and concepts, so you can zoom into the most relevant components.

Quick Start

Ask for an explanation of a specific location by saying: /explain src/auth/middleware.ts

Frequently Asked Questions about explain

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

FAQPage Schema
How do I explain unfamiliar code and understand how it fits into the surrounding system?

To explain unfamiliar code, this Skill produces structured, layered breakdowns of file paths, symbols, or concepts. It clarifies purpose, relationships, and where to safely make changes by gathering focused context and analyzing callers and dependencies.

What is the best way to navigate a large codebase and understand its architecture?

Navigating a large codebase is best handled by generating a map-first overview for directories and concepts. This approach zooms into the most relevant components and grounds explanations using the repository's CLAUDE.md Architecture section to match team conventions.

How do I analyze code dependencies and callers before modifying a function?

Analyzing code dependencies and callers is done by parsing the user input type and gathering limited focused context. This identifies surrounding connections and entry points, producing a six-section explanation to clarify safe modification areas.

Does this code explanation approach work with both single functions and broader directory scopes?

Yes, code explanation works with single functions, files, directories, and natural-language concepts. It uses adaptive depth to choose the right level of detail based on what you are pointing at, providing map-first overviews for wider directory scopes.

What conventions and gotchas are included when explaining a codebase?

Explaining a codebase includes a six-section breakdown covering patterns, conventions, and gotchas. It grounds these insights in the project's CLAUDE.md Architecture section when available, matching team terminology and highlighting potential issues.