Back to Skills

content-refiner

majiayu000
Updated Yesterday
58
9
58
View on GitHub
Documentsword

About

The content-refiner skill streamlines verbose educational content by removing redundancy and tightening prose to improve narrative flow. It is triggered when content is wordy, repetitive, or has disconnected lessons. Key actions include consolidating concepts, eliminating unnecessary sections, and strengthening connections between lesson parts.

Quick Install

Claude Code

Recommended
Plugin CommandRecommended
/plugin add https://github.com/majiayu000/claude-skill-registry
Git CloneAlternative
git clone https://github.com/majiayu000/claude-skill-registry.git ~/.claude/skills/content-refiner

Copy and paste this command in Claude Code to install this skill

Documentation

Content Refiner Skill

Transform verbose, redundant educational content into lean, connected lessons.

When to Use

  • Content feedback mentions: verbose, redundant, wordy, repetitive
  • Lessons feel disconnected or read like standalone blog posts
  • Same concept explained multiple ways within a lesson
  • "Try With AI" sections have 4+ prompts
  • Lessons exceed 1200 words without justification

The Three Enemies

Enemy 1: Verbosity

Symptoms:

  • Multiple analogies for the same concept
  • "Why This Matters" sections that restate the obvious
  • Tables that duplicate paragraph content
  • "Reflection" sections that add no value

Treatment:

  • ONE analogy per concept maximum
  • Cut "Why This Matters" unless it reveals non-obvious insight
  • Choose: paragraph OR table, not both
  • Delete "Reflection" sections entirely

Enemy 2: Redundancy

Symptoms:

  • Concept explained in Lesson N, re-explained in Lesson N+1
  • Same information in different formats (paragraph, bullets, table)
  • "Expert Insight" callouts restating what was just said
  • Multiple lessons that could be one

Treatment:

  • Concepts taught ONCE, referenced thereafter
  • One format per concept
  • Expert Insights only for genuinely advanced perspectives
  • Merge lessons that cover same ground

Enemy 3: Disconnection

Symptoms:

  • Each lesson reads like standalone article
  • No "Previously you learned X, now we build on Y" bridges
  • Different examples in each lesson (no running example)
  • Conceptual lessons sandwiched between practical ones

Treatment:

  • Opening sentence references prior lesson's key takeaway
  • ONE running example evolves across the chapter
  • Conceptual content folded INTO practical lessons
  • Clear skill progression: each lesson adds ONE new capability

Refinement Procedure

Step 1: Measure Current State

Count:
- Total words
- Number of analogies per concept
- Number of "Try With AI" prompts
- Number of tables
- "Reflection" sections present?
- "Expert Insight" callouts

Step 2: Apply Cuts

Mandatory cuts:

  1. Delete ALL "Reflection" sections
  2. Reduce "Try With AI" to exactly 2 prompts
  3. Keep ONE analogy per concept, delete others
  4. Delete tables that duplicate paragraph content
  5. Cut "Why This Matters" if it only restates the concept

Word targets:

Lesson TypeTarget Words
Conceptual intro600-800
Hands-on practical800-1000
Installation/setup400-600
Capstone1000-1200

Step 3: Strengthen Connections

Opening formula:

# [Lesson Title]

In [Lesson N-1], you [key accomplishment]. Now you'll [this lesson's goal].

Running example rule:

  • Identify the chapter's running example
  • This lesson MUST use or extend that example
  • If introducing new example, it must relate to running example

Step 4: Verify Quality

Checklist:

  • Under word limit for lesson type
  • One analogy per concept max
  • Exactly 2 "Try With AI" prompts
  • No "Reflection" section
  • Opens with connection to prior lesson
  • Uses or extends running example
  • No repeated explanations from earlier lessons

Output Format

When refining a lesson, produce:

## Refinement Report: [Lesson Name]

### Metrics
| Before | After |
|--------|-------|
| X words | Y words |
| N analogies | 1 analogy |
| N Try With AI | 2 prompts |

### Key Cuts Made
1. [Deleted section/content and why]
2. [Deleted section/content and why]
3. [Deleted section/content and why]

### Connection Added
- Opening: "[New opening sentence]"
- Running example: [How it connects]

### Refined Content
[Full refined lesson content]

Example: Before/After

BEFORE (verbose):

## Why This Matters

Skills are important because they save you time. When you create a skill,
you're investing once to benefit forever. Think of it like teaching a
friend your preferences. Or like programming a robot. Or like writing
a recipe book. The key insight is that skills encode your expertise.

| Aspect | Without Skills | With Skills |
|--------|---------------|-------------|
| Time | Repeat yourself | Invest once |
| Quality | Inconsistent | Consistent |
| Sharing | Hard | Easy |

As you can see, skills provide significant advantages...

AFTER (lean):

Skills encode your expertise once so Claude applies it automatically.
Instead of explaining your LinkedIn tone every time, teach it once.

Anti-Patterns to Eliminate

  1. The Triple Explanation: Paragraph + Table + Analogy for same concept
  2. The Standalone Syndrome: Lesson that doesn't reference what came before
  3. The Prompt Explosion: 4+ "Try With AI" prompts
  4. The Obvious Insight: "Expert Insight" that adds nothing experts wouldn't know
  5. The Setup Novel: 3 paragraphs of motivation before getting to content
  6. The Example Carousel: New example every lesson instead of building one

Skill Composition

This skill works well with:

  • content-implementer: Apply these principles when creating new content
  • educational-validator: Validate refined content still meets pedagogical requirements

GitHub Repository

majiayu000/claude-skill-registry
Path: skills/content-refiner

Related Skills

content-collections

Meta

This skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.

View skill

cloudflare-turnstile

Meta

This skill provides comprehensive guidance for implementing Cloudflare Turnstile as a CAPTCHA-alternative bot protection system. It covers integration for forms, login pages, API endpoints, and frameworks like React/Next.js/Hono, while handling invisible challenges that maintain user experience. Use it when migrating from reCAPTCHA, debugging error codes, or implementing token validation and E2E tests.

View skill

llamaindex

Meta

LlamaIndex is a data framework for building RAG-powered LLM applications, specializing in document ingestion, indexing, and querying. It provides key features like vector indices, query engines, and agents, and supports over 300 data connectors. Use it for document Q&A, chatbots, and knowledge retrieval when building data-centric applications.

View skill

canvas-design

Meta

The canvas-design skill generates original visual art in PNG and PDF formats for creating posters, designs, and other static artwork. It operates through a two-step process: first creating a design philosophy document, then visually expressing it on a canvas. The skill focuses on original compositions using form, color, and space while avoiding copyright infringement by never copying existing artists' work.

View skill