amhuppert/my-ai-resources/explainer-document

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Write a concise explainer / learning document / beginner guide that

Install

npx vskill@latest install amhuppert/my-ai-resources/explainer-document

SKILL.md

Explainer Document

Create concise, beginner-friendly educational documents that teach a topic through logical progression. The audience is someone who does not already understand the subject.

Process

1. Define Scope

Before writing, establish:

  • Topic boundaries — what is and isn't covered
  • Target audience — what prerequisite knowledge can be assumed
  • Depth — how deep to go (overview vs. working knowledge)

If any of these are ambiguous, ask before proceeding.

2. Identify Core Concepts

List every concept the reader needs to understand. Then determine the dependency order: which concepts require understanding other concepts first.

Example dependency chain:

"Permission" depends on "API" → "Role" depends on "Permission" → "Client Grant" depends on both "Application" and "API"

This dependency order determines the document's section order.

3. Research and Verify

For technical topics, verify claims against official documentation. Use web search and official docs — do not rely solely on training data. If something cannot be verified, say so explicitly rather than stating it as fact.

4. Write the Document

Follow the structure and style guidelines below.

5. Review

After writing, check:

  • Can each section be understood using only the sections before it?
  • Are there terms used before they're defined?
  • Is anything included that a beginner doesn't need yet?
  • Is anything essential missing?

Document Structure

Opening

Start with the most fundamental concepts — the building blocks everything else depends on. Define key entities and their relationships before discussing behaviors or mechanics.

Progression

Each section builds on the previous. The reader should never encounter an unexplained term. Structure from:

  1. What things are (entities, definitions)
  2. How things relate (relationships, interactions)
  3. How things behave (mechanics, rules, edge cases)
  4. Nuances and caveats (non-obvious behavior, common misconceptions)
Section Design
  • Each section covers one concept or closely related group
  • Keep sections short — a reader should be able to scan the heading and get the gist
  • Use descriptive headings that communicate content (not generic labels like "Details" or "More Information")

Style Guidelines

Voice and Tone
  • Declarative and direct — state how things work, not how they might work
  • No hedging or filler ("It should be noted that...", "It's worth mentioning...")
  • Confident but honest — if something is uncertain or context-dependent, say so plainly
Formatting for Scannability
  • Bold key terms on first introduction
  • Use tables for comparisons, multi-dimensional relationships, and quick-reference summaries
  • Use bulleted lists for sets of items or properties
  • Use numbered lists only for sequential steps or ranked items
  • Use headings liberally — they are the primary navigation tool for scanners
Conciseness
  • Cover the essentials without being exhaustive
  • Every sentence should teach something — remove sentences that only connect or transition
  • Prefer a concrete example over an abstract explanation when both would take similar space
  • Do not repeat information across sections; reference earlier sections if needed
Defining Terms
  • Define terms where they first appear, inline with the explanation
  • Do not create a separate glossary — definitions should be encountered in context
  • Use the format: Term — definition. (Bold term, em dash, explanation.)

Example:

Application — an OAuth 2.0 client that requests tokens. Identified by a client_id and client_secret.
Tables

Use tables when comparing across multiple dimensions. Tables are especially effective for:

  • Showing how behavior differs across contexts or configurations
  • Summarizing "if X then Y" relationships
  • Quick-reference summaries of a section's key points
Critical Distinctions

When the topic has a commonly confused or non-obvious behavior, call it out with a clearly labeled subsection (e.g., "Critical: These Are Mutually Exclusive"). Do not bury important caveats in the middle of a paragraph.

Anti-Patterns

  • Wall of text — if a section has more than ~6 lines of prose without a heading, list, or table, break it up
  • Jargon without definition — every domain term must be defined before or at the point of use
  • Backwards dependencies — never reference a concept that hasn't been introduced yet
  • Exhaustive coverage — this is a learning document, not a reference manual; cover what's needed to understand the topic, not every edge case
  • Condescending tone — the reader is a beginner to this topic, not to thinking; avoid phrases like "simply", "just", "of course"

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/amhuppert/my-ai-resources/explainer-document/badge)](https://verified-skill.com/skills/amhuppert/my-ai-resources/explainer-document)

Something wrong with this skill or its page? Tell us.