generate-explanations

Write and review Diátaxis explanation documents with structure, voice, and checklist rules.

Updated Sep 7, 2026
One-click install
npx skills add https://github.com/deriye/.agents-work --skill generate-explanations-deriye
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: generate-explanations
Source: https://github.com/deriye/.agents-work/tree/main/skills/generate-explanations
Command: npx skills add https://github.com/deriye/.agents-work --skill generate-explanations-deriye

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Documentation often blurs tutorials, how-tos, and reference material into conceptual writing, producing pages that fail readers who want context and understanding rather than steps. This Skill enforces the Diátaxis explanation mode so conceptual docs stay discursive, bounded, and opinionated. ## Core Features & Use Cases - Mode classification: Distinguishes explanation from tutorials, how-to guides, and reference material before writing begins. - Structure and voice guidance: Provides opening, body, and closing patterns plus a discursive language table (why, however, some prefer) that keeps imperative instruction out. - Publish checklist: A 12-point author checklist gates publication, covering bounded scope, context, connections, alternatives, and reflection-oriented writing. - Use Case: When asked to write an "About retrieval-augmented generation" page, the Skill produces a discussion of why RAG exists, its trade-offs, and alternatives, while linking out to tutorials and reference instead of absorbing them. ## Quick Start Ask the AI to write or review an explanation document about a specific topic, such as "write an explanation about why we chunk documents in RAG".

Frequently Asked Questions about generate-explanations

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

FAQPage Schema
How do I write a Diátaxis explanation document?▼

Start from a why question or a drawn boundary, then write discursive prose covering context, history, trade-offs, and alternatives. Open with what the piece is about, circle the subject from multiple angles, and close by restating the picture without adding steps or API catalogues.

What is the difference between explanation and reference documentation?▼

Explanation is discursive discussion read away from the product to build understanding, while reference is neutral, complete description consulted during a task. Explanation allows opinion and perspective; reference does not. The test is study versus work, not complexity.

When should I not use the explanation writing mode?▼

Do not use it for tutorials with numbered learning steps, how-to guides for real-world tasks, or reference catalogues of APIs, flags, and schemas. If the reader would open the page while executing a task, it is not explanation.

Why does my explanation document read like a tutorial?▼

It likely uses imperative do-voice, numbered procedures, or learning-outcome framing like "you will learn". Replace steps with discussion of reasons and trade-offs, frame the title as About a topic, and move any procedures into a linked tutorial.

Can an explanation include code examples or commands?▼

Yes, but only as illustration of a claim, trade-off, or historical practice, never as a path the reader must reproduce. If removing the run-this implication collapses the section, the content is instruction and belongs in a tutorial or how-to.