Przejdź do treści
Biblioteka skilli
Dokumenty i biuro · Pisanie

proof-read

/proof-read

Poprawia tekst pod kątem jasności, zwięzłości i naturalnego przepływu, zwłaszcza treści brzmiące sztucznie, rozwlekle lub powtarzalnie.

Repozytorium
skills ↗
Licencja
MIT
Ostatnia zmiana repozytorium
28.10.2026

Ocena i bezpieczeństwo

100/ 100Klasa: dobry
Format25 / 25
Frontmatter z nazwą i opisem, poprawna nazwa, opis do 1024 znaków, nazwa zgodna z folderem.
Odkrywalność25 / 25
Opis mówi, kiedy użyć skilla, ma sensowną długość i jest w 3. osobie.
Treść35 / 35
Co najmniej 150 słów i do 500 linii, nagłówki, listy kroków, przykłady albo kod.
Higiena15 / 15
Bez zaślepek typu TODO, bez ścieżek Windows, z licencją.
Ryzykobrak sygnałów w skanie
Pliki w paczce
1
Skrypty
0
Komentarze HTML
0
Adresy URL w SKILL.md
0
Przestarzałość

Brak sygnałów. Repozytorium zmienione 28.10.2026, bez odwołań do wycofanych modeli i API.

Licencja

MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.

Co robi i kiedy

Poprawia tekst pod kątem jasności, zwięzłości i naturalnego przepływu, zwłaszcza treści brzmiące sztucznie, rozwlekle lub powtarzalnie.

Oryginalny opis (EN)

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.

Jak działa

  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
Sekcje z nagłówków SKILL.md, w oryginale · 151 linii, 1 002 słów
Podgląd SKILL.md
---
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"

Wersje

W bazie nie ma innych skilli o nazwie proof-read.

Migawka bazy z 13.11.2026 · wersja paczki 8e403e8fe1bb (skrót SHA-256 całej paczki)

Pobrania w ostatnich 30 dniach: 0 · ranking

Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.

Podobne skille

  • 90Prompt Brevity Review
    Recenzuje prompty i instrukcje AI pod kątem zwięzłości i jasności, w tym skille, CLAUDE.md i komendy slash.
  • 100copy-editing
    Redaguje linia po linii istniejące teksty marketingowe pod kątem jasności, zwięzłości, siły przekazu i konwersji; do pisania od zera służy copywriting.
  • 100unslop
    Usuwa ślady AI z tekstów pisanych dla człowieka i poprawia brzmienie, gdy tekst wygląda na wygenerowany.
  • 85red-pen
    Redaguje tekst pod kątem jasności, zwięzłości i typografii, czyli korekta liniowa bez zmiany sensu.
  • 100content-humanizer
    Przekształca teksty wygenerowane przez AI w naturalniejsze, wykrywając charakterystyczne wzorce, przywracając rytm zdań i oceniając autentyczność, gdy tekst brzmi sztucznie.

Uwagi i zgłoszenia

Wiadomość trafi do Bartosza Sokolińskiego. Adres e-mail jest potrzebny tylko do odpowiedzi. Szczegóły w polityce prywatności.