anjijava16/agent_skills/casualize-names

Casualize Names Workflow

Goal

Convert formal names (first names, company names, cities) to casual, friendly versions suitable for cold email copy.

Scripts

• ./scripts/casualize_batch.py - Main script (all 3 fields at once)
• ./scripts/casualize_company_names_batch.py - Company names only
• ./scripts/casualize_first_names_batch.py - First names only
• ./scripts/casualize_city_names_batch.py - City names only

Quick Start

# Process all three fields at once (recommended, 3x faster)
python3 -u ./scripts/casualize_batch.py "GOOGLE_SHEET_URL"
 
# Re-process existing (overwrite)
python3 -u ./scripts/casualize_batch.py "GOOGLE_SHEET_URL" --overwrite

How It Works

1. Processes records in batches of 50
2. Uses 5 parallel workers
3. Claude converts all three fields in one API call
4. Batch updates Google Sheet with results
5. Only processes rows with emails

Performance: ~35 records/sec (3,000 records ≈ 90 seconds)

Casualization Rules

First Names

• Use common nicknames: "William" → "Will", "Jennifer" → "Jen"
• Keep original if no common nickname exists
• Keep it professional

Company Names

• Remove "The" at beginning
• Remove legal suffixes (LLC, Inc, Corp, Ltd)
• Remove generic words (Realty, Group, Solutions, Services)
• Keep core brand name
• Use "you guys" for overly generic names

Examples:

• "Keller Williams Realty Inc" → "Keller Williams"
• "The Teal Umbrella Family Dental Healthcare" → "Teal Umbrella"

City Names

• Use local nicknames: "San Francisco" → "SF", "Philadelphia" → "Philly"
• Keep original if no common nickname

Output

Creates three new columns:

• casual_first_name
• casual_company_name
• casual_city_name

Environment

ANTHROPIC_API_KEY=your_key