playbook-ai-specificity

Generates a company-specific offer sentence tail as a cold email merge field.

678|243|Updated Mar 10, 2026
One-click install
npx skills add https://github.com/growthenginenowoslawski/coldoutboundskills --skill playbook-ai-specificity
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: playbook-ai-specificity
Source: https://github.com/growthenginenowoslawski/coldoutboundskills/tree/main/skills/playbooks/playbook-ai-specificity
Command: npx skills add https://github.com/growthenginenowoslawski/coldoutboundskills --skill playbook-ai-specificity

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Cold emails with a good offer still read like they were sent to 5,000 people because the offer is stated generically. This playbook writes the one sentence tail that restates the client's capability in the prospect's own nouns, so the offer reads as written for that one company.

Core Features & Use Cases

  • Specificity line generation: Produces a 6-14 word, lowercase, verb-first tail that slots into the fixed frame "I think we can help you {{specificity_line}}.", with blank as a valid output for non-fit companies.
  • Deterministic verification guard: Mechanically checks that the anchor is quoted from the company evidence, the offer item is quoted from the locked client offer block, and the line passes length, dash, capitalization, and banned-phrase rules.
  • Reading-level gating: Enforces Flesch-Kincaid grade 7 on the generated tail only, since the fixed frame alone scores 5.7 and gating the rendered sentence would blank every row.
  • Use Case: A bookkeeping firm emailing ecommerce brands replaces "we help with margin reporting" with "know the landed cost of every pool float before you price it" for a pool supply company, generated per lead from its company description.

Quick Start

Ask Claude to run the playbook-ai-specificity skill on a lead list with a locked client offer block to generate a specificity_line merge field for each company.

Frequently Asked Questions about playbook-ai-specificity

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a personalized cold email line for each company?

Generate a specificity line: a 6-14 word tail that slots into the fixed frame "I think we can help you {{specificity_line}}." It names what the client would do using a real product noun copied from the company's own description, rather than a generic capability statement.

How do I stop AI-generated personalization lines from sounding generic?

Force the model to quote an anchor word from the company data and an offer item from a locked client offer list, then verify both quotes deterministically. Banned-phrase rules alone just move fluff around; the anchor check is what blocks invented or vague lines.

Should I gate the reading level of the full rendered email sentence?

No, gate only the generated tail. The frame "Specifically, I think we can help you." scores 5.7 on Flesch-Kincaid by itself, so any 6-14 word tail renders at 7.0-9.8 regardless, and gating the render blanks every row.

What happens when a company does not fit the offer?

The line returns blank, which is a valid answer covering roughly 10-20% of a normal list. Route abstains out of the campaign or pre-render the whole sentence into one field; spintax cannot branch on empty variables.

Why does fetching company homepages fail for personalization evidence?

Plain homepage fetches on modern ecommerce sites frequently return nothing; measured results included 0 bytes, 16 bytes, 1 byte, and refused connections. Use a derived company description first and a capped rendering proxy only as a fallback.