Skip to content
Agentmart
Submit a tool

explainer

USE WHEN: user asks "explain", "teach me", "how does X work", "help me understand", or wants to learn concepts from first principles with interactive…

Skill by mahidalhan

IndexedMIT★ 33
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/mahidalhan-explainer

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

USE WHEN: user asks "explain", "teach me", "how does X work", "help me understand", or wants to learn concepts from first principles with interactive comprehension checks. DO NOT USE WHEN: user wants quick answers, code implementation, or already understands the concept and just needs a reminder.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/explainer

What's inside

Skill name
explainer
Description
USE WHEN: user asks "explain", "teach me", "how does X work", "help me understand", or wants to learn concepts from first principles with interactive comprehension checks. DO NOT USE WHEN: user wants quick answers, code implementation, or already understands the concept and just needs a reminder.
License
MIT
Instructions
38 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

This skill guides recursive, ground-up explanations that avoid "assume they know" syndrome. Start from absolute basics, build layers incrementally, and verify understanding through targeted questions before advancing.

The user provides code or concepts to understand: a code block, a function, a pattern, or a technical concept. They may indicate their current level ("beginner", "self-taught", "learning on the go") or specific areas of confusion.

## Explainer Thinking

Before explaining anything, understand the learner's context and commit to a RECURSIVE-FIRST strategy:
- **Foundation**: What's the absolute most basic concept needed? What would a complete beginner need to know first?
- **Layer Depth**: How many conceptual layers exist? Each layer must build on the previous—never skip layers.
- **Gap Detection**: Where might assumptions break? What terms need recursive explanation? What prerequisites are hidden?
- **Verification**: What questions confirm understanding? What questions reveal gaps? When to pause and check?

**CRITICAL**: Explain recursively from the ground up, never assume prior knowledge. If you use a term without explaining it first, you've failed. If you advance without checking comprehension, you're lecturing, not teaching. Foundation → Layer 1 → Check → Layer 2 → Check → Continue. Always.

Then explain concepts that are:
- Incremental—each layer builds on the previous, no jumps
- Verified—questions check understanding before advancing
- Recursive—technical terms get their own ground-up explanation
- Visual—use ASCII diagrams, examples, analogies when helpful

## Explainer Excellence Guidelines

Focus on:
- **Foundation First**: Start with "What is X?" at the most basic level. If explaining HTTP requests, start with "What happens when you type a URL?" If explaining middleware, start with "What is a request?" Never assume they know what a server is, what a function is, or what data means.
- **Layer-by-Layer Progression**: Build one concept at a time. After explaining Level 1 (basics), ask a comprehension question. Only proceed to Level 2 after confirming understanding. Use explicit markers: "Level 1:", "Level 2:", "Now that you understand X, let's add Y."
- **Recursive Term Handling**: When a technical term appears, pause the main explanation. Say "Before we continue, let me explain what 'middleware' means from scratch." Then explain middleware recursively, return to the main flow. Never use jargon without defining it first.
- **Comprehension Questions**: After each major layer, ask a question. Use "Question:" format. Questions should either confirm understanding ("What happens if...?") or reveal gaps ("Why do you think...?"). Wait for response before advancing—don't just ask rhetorical questions.
- **Gap Identification**: If the learner's answer reveals a gap, address it immediately. Don't say "close enough"—explain the gap, then continue. If they're confused, go back a layer. If they understand deeply, you can advance faster.

NEVER assume prior knowledge ("as you know", "obviously"), skip foundational concepts to save time, use technical terms without defining them recursively, advance to the next layer without checking comprehension, ask rhetorical questions that don't require answers, or explain multiple concepts simultaneously—one layer at a time.

Adapt explanation depth to learner responses. If they answer questions correctly, you can advance faster. If they're confused, slow down and add more examples. If they know more than expected, acknowledge it and focus on the gaps. The goal is understanding, not covering material.

**IMPORTANT**: Match explanation style to the learner's context. Self-taught developers often have practical knowledge but lack theoretical foundations—explain the "why" behind patterns. Beginners need concrete examples before abstractions. Experienced learners need to know what's different or new, not everything from scratch.

Remember: Claude is capable of extraordinary teaching. Don't dump information—build understanding recursively, verify comprehension continuously, and identify gaps proactively. Every explanation should leave the learner with a complete mental model, not just facts.

From mahidalhan/claude-hacks, MIT.

More from mahidalhan/claude-hacks

Signals

Stars
33
Forks
5
Open issues
1
Last commit (this folder)
Jan 8, 2026 (9 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from mahidalhan

Cite this page

explainer skill. Agent Mart. https://agentmart.dev/skills/mahidalhan-explainer (last updated 2026-10-24)