Skip to content
Skills Library
Documents & office · Writing

proof-read

/proof-read

Use when refining text for clarity, conciseness, and natural flow. Use when editing AI-generated content that sounds robotic, wordy, or repetitive. Use when finalizing any text for publication where polish and human authenticity matter.

Repository
skills ↗
Licence
MIT
Last repository change
28.10.2026

Assessment and security

100/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 28.10.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Use when refining text for clarity, conciseness, and natural flow. Use when editing AI-generated content that sounds robotic, wordy, or repetitive. Use when finalizing any text for publication where polish and human authenticity matter.

Original author description

Use when refining text for clarity, conciseness, and natural flow. Use when editing AI-generated content that sounds robotic, wordy, or repetitive. Use when finalizing any text for publication where polish and human authenticity matter.

How it works

  1. Overview
  2. When to Use
  3. The Proof-Read Process
  4. Three-Pass System
  5. AI-Generated Text Markers
  6. Deliverables
  7. Common Mistakes
  8. Red Flags — STOP and Reassess
  9. Example Transformation
Original SKILL.md headings · 151 lines, 1,002 words
SKILL.md preview
---
name: proof-read
description: Use when refining text for clarity, conciseness, and natural flow. Use when editing AI-generated content that sounds robotic, wordy, or repetitive. Use when finalizing any text for publication where polish and human authenticity matter.
---

# Proof-Read

## Overview

**Proof-reading is the final refinement stage that transforms rough drafts into polished, human-sounding text.**

This skill combines mechanical accuracy (grammar, spelling, punctuation) with editorial judgment (clarity, flow, tone). The goal is not just correctness—it's creating text that reads naturally and engages the reader without friction.

## When to Use

- Final pass before publishing or sending any written content
- Editing AI-generated text that feels robotic or generic
- Trimming wordy drafts to essential message
- Unifying tone and style across multi-author documents
- Fixing "almost right" text that isn't quite landing

**Do NOT use for:**
- Initial drafting — use outlining or brainstorming skills first
- Major structural rewrites — reorganize content before proof-reading
- Fact-checking research claims — use text-assessment for verification
- Style guide compliance (APA, Chicago, etc.) — refer to specific guides

## The Proof-Read Process

```dot
digraph proofread_flow {
    "First Pass: Mechanical" [shape=box];
    "Second Pass: Editorial" [shape=box];
    "Third Pass: Humanization" [shape=box];
    "Final Check: Read Aloud" [shape=diamond];
    "Deliver with Summary" [shape=box];
    
    "First Pass: Mechanical" -> "Second Pass: Editorial";
    "Second Pass: Editorial" -> "Third Pass: Humanization";
    "Third Pass: Humanization" -> "Final Check: Read Aloud";
    "Final Check: Read Aloud" -> "Third Pass: Humanization" [label="sounds off"];
    "Final Check: Read Aloud" -> "Deliver with Summary" [label="flows naturally"];
}
```

## Three-Pass System

### Pass 1: Mechanical (Accuracy)

Check for objective errors:

| Check | What to Look For |
|-------|------------------|
| **Spelling** | Typos, homophones (their/there/they're), proper nouns |
| **Grammar** | Subject-verb agreement, tense consistency, article usage |
| **Punctuation** | Comma splices, missing commas, incorrect apostrophes |
| **Facts** | Dates, names, numbers, statistics (flag if unverifiable) |
| **Consistency** | Capitalization, hyphenation, formatting, tense |

### Pass 2: Editorial (Clarity & Impact)

Refine for reader experience:

| Issue | Fix Strategy |
|-------|--------------|
| **Wordiness** | Cut "in order to" → "to"; "due to the fact that" → "because" |
| **Repetition** | Merge similar sentences; use pronouns or synonyms |
| **Complex sentences** | Split >25 word sentences; one idea per sentence |
| **Weak openings** | Start with subject + verb; avoid "There is/are" |
| **Jargon** | Replace insider terms with plain language or define them |
| **Passive voice** | Convert to active where it clarifies who acts |

### Pass 3: Humanization (Tone & Flow)

Remove robotic patterns common in AI output:

| AI Pattern | Human Alternative |
|------------|-------------------|
| "It is important to note that..." | Delete or replace with direct statement |
| "In today's world..." / "In the modern era..." | Delete — get to the point |
| "This comprehensive analysis will..." | State what it does directly |
| Lists of three where two suffice | Cut the weakest item |
| "Leverage," "utilize," "facilitate" | Use "use," "help," "make" |
| Overly formal transitions | Use conversational connectors |
| Perfect parallel structure | Vary sentence rhythm slightly |
| "Not only X but also Y" constructions | Simplify to "X and Y" |

## AI-Generated Text Markers

Flag these common AI tells:

- **Hedging language**: "It is worth noting," "One might consider," "It seems that"
- **Generic intensifiers**: "significant," "crucial," "vital," "essential" without specificity
- **Bullet point overuse**: Three-item lists where prose flows better
- **Abstract openings**: Starting with definitions or broad context before the point
- **Conclusion restatements**: Summarizing what was just said in the final paragraph
- **Even rhythm**: Sentences of similar length creating a metronome effect

## Deliverables

After editing, provide:

1. **Edited text** — The polished version ready for use

2. **Summary of changes** including:
   - Word count reduction (before → after)
   - Major restructuring changes
   - Specific AI-like phrases humanized
   - Consistency fixes applied

### Summary Template

```
Changes Made:
• Word count: [X] → [Y] ([Z]% reduction)
• Mechanical: [N] spelling/grammar/punctuation fixes
• Editorial: [Specific changes, e.g., "Split 4 long sentences"]
• Humanization: [Specific AI patterns removed, e.g., "Removed 'It is important to note' hedging"]
• Flow improvements: [e.g., "Varied sentence rhythm, strengthened paragraph openings"]
```

## Common Mistakes

| Mistake | Why It Fails | Fix |
|---------|--------------|-----|
| Editing in one pass | Misses issues that need distance | Take breaks between passes |
| Preserving author's voice when it sucks | Polished mediocrity is still mediocre | Fix structural issues first |
| Over-editing | Removes personality and warmth | Stop when it sounds natural |
| Only reading silently | Misses awkward rhythm and missing words | Read final version aloud |
| Changing meaning while tightening | Accuracy trumps brevity | Verify facts after cuts |
| Ignoring context | What's clear to you may not be to reader | Consider audience knowledge |

## Red Flags — STOP and Reassess

- **Word count increased**: Proof-reading should reduce or maintain length
- **Lost the main point**: If the core message got buried, undo and restart
- **Sounds like a different person**: Too much editing strips authentic voice
- **Can't read aloud smoothly**: If you stumble reading it, readers will too
- **Multiple passes, still not right**: The text may need rewriting, not editing

## Example Transformation

**Original (AI-like)**:
> It is important to note that in today's fast-paced business environment, organizations must leverage comprehensive data analytics solutions in order to remain competitive. Not only do these tools facilitate better decision-making processes, but they also enable teams to identify significant trends and patterns. It is worth mentioning that companies that fail to adopt these technologies may face substantial challenges.

**Edited (Human)**:
> Organizations need data analytics to stay competitive. These tools help teams spot trends and make better decisions. Companies that ignore them fall behind.

**Changes**:
- Word count: 57 → 28 (51% reduction)
- Removed hedging: "It is important to note," "It is worth mentioning"
- Removed filler: "in today's fast-paced business environment," "comprehensive," "in order to"
- Simplified: "leverage" → "need", "facilitate" → "help", "significant trends and patterns" → "trends"
- Strengthened ending: "may face substantial challenges" → "fall behind"

Versions

The database has no other skills named proof-read.

Database snapshot: 13.11.2026 · package version 8e403e8fe1bb (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 90Prompt Brevity Review
    Review AI prompts and instructions for conciseness and clarity. Use when reviewing skills, CLAUDE.md, slash commands, or any LLM prompt content.
  • 100copy-editing
    ACTIVATE when the user has EXISTING marketing copy and wants it polished — line-by-line editing for clarity, conciseness, impact, conversion. Trigger phrases: 'edit my copy', 'improve this copy', 'polish this text', 'copy edit', 'make this clearer', 'tighten this up',…
  • 100unslop
    Use when writing or editing prose a person will read, or when text sounds AI-generated.
  • 85red-pen
    Line-edit prose for clarity, economy and typography. Use for 'red pen', 'red-pen this', 'copyedit', 'line edit', 'proofread' or 'tighten this prose'. For making text sound less machine-written, use humanize.
  • 100content-humanizer
    Transform AI-generated content into human-sounding writing via AI pattern detection, rhythm restoration, and authenticity scoring. Use when content sounds robotic, uses AI cliches, or the user wants to humanize or fix AI writing.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.