humanize.py

#!/usr/bin/env python3
"""Remove AI writing tells from text.

Common giveaways:
- Em dashes (—) → regular dash (-)
- "Delve/dive into" → "look at"
- "It's worth noting" → delete
- "Importantly" → delete
- Perfect parallel structure → roughen it up
- Overly clean lists → add variance
"""

import re
import sys

AI_TELLS = {
    # Em dashes
    r'—': '-',
    
    # Overused phrases
    r'\bdelve into\b': 'look at',
    r'\bdive into\b': 'look at',
    r'\bunpack\b': 'examine',
    r"It's worth noting that\s*": '',
    r"Let's be clear:\s*": '',
    r"Let's be honest:\s*": '',
    r'\bImportantly,\s*': '',
    r'\bCrucially,\s*': '',
    r'\bFundamentally,\s*': '',
    
    # Sentence starters
    r'\bThis is where\b': 'Here',
    r'\bThat being said,\s*': '',
    r'\bWith that in mind,\s*': '',
    
    # Hedging
    r'\bquite\b': '',
    r'\bsomewhat\b': '',
    r'\brather\b': '',
    r'\ba bit\b': '',
}

def humanize(text):
    """Apply all transformations."""
    result = text
    for pattern, replacement in AI_TELLS.items():
        result = re.sub(pattern, replacement, result, flags=re.IGNORECASE)
    
    # Clean up double spaces
    result = re.sub(r'  +', ' ', result)
    
    # Clean up space before punctuation
    result = re.sub(r' +([.,!?])', r'\1', result)
    
    return result.strip()

def detect_tells(text):
    """Find AI tells in text."""
    tells = []
    
    if '—' in text:
        tells.append(f"Em dash found: {text.count('—')} instances")
    
    for pattern in AI_TELLS.keys():
        if re.search(pattern, text, re.IGNORECASE):
            tells.append(f"AI phrase: {pattern}")
    
    # Check for perfect parallel structure (3+ sentences starting with same word)
    sentences = [s.strip() for s in re.split(r'[.!?]', text) if s.strip()]
    if len(sentences) >= 3:
        first_words = [s.split()[0] if s.split() else '' for s in sentences]
        for word in set(first_words):
            if first_words.count(word) >= 3:
                tells.append(f"Parallel structure: {first_words.count(word)} sentences start with '{word}'")
    
    return tells

if __name__ == '__main__':
    if len(sys.argv) > 1:
        text = ' '.join(sys.argv[1:])
    else:
        text = sys.stdin.read()
    
    print("=== AI TELLS DETECTED ===")
    tells = detect_tells(text)
    if tells:
        for tell in tells:
            print(f"  • {tell}")
    else:
        print("  None found")
    
    print("\n=== HUMANIZED VERSION ===")
    print(humanize(text))