What problem does it solve? Writing personalized cold email lines for hundreds of leads by hand is slow, and running a single prompt over a large list without review produces generic or wrong copy. This Skill defines a reusable approval-loop pattern that tunes a personalization prompt on small batches before scaling to the full lead list. ## Core Features & Use Cases - Approval loop before scale: Shows one sample personalization, then batches of 10, collecting user edits until two consecutive rounds pass with zero corrections, then locks the prompt. - Parallel Task sub-agent fan-out: Splits leads into batches of 10-20 and launches multiple Claude Code Task sub-agents (including A/B/C copy variants) that write JSON results to /tmp for merging by lead_id. - Schema and quality guardrails: Enforces a strict output schema (situation_line, value_line, cta_soft), banned-word rules, no-fabrication rules, skip handling for thin data, and spot-check guidance before upload to Smartlead or Instantly. - Use Case: Given 100 enriched leads with company descriptions, tune the prompt on 20 leads across two approved rounds, then fan out 10 parallel sub-agents to generate three personalization variants per lead and map them to custom variables in your sending platform. ## Quick Start Ask Claude to personalize this batch of leads using the personalization sub-agent pattern, starting with one sample lead for approval before fanning out.
