list-expander

Expand seed companies into a qualified TAM via lookalikes, mined filters, and AI verification.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Niche B2B lists come out tiny because database keyword searches miss good-fit companies that aren't tagged with the obvious terms. This Skill works bottom-up from ~10 known-good seed companies to discover how databases actually tag them, expand via lookalike engines, and build a full qualified TAM with measured precision.

Core Features & Use Cases

  • Seed Fingerprinting & Lookalike Generation: Records how each seed appears in Prospeo plus its live homepage, then expands via Prospeo company_lookalike, Exa findSimilar, and Parallel.ai entity search.
  • Filter Mining with Precision Scorecard: Mines industries and keyword n-grams from confirmed fits, scores each candidate filter on volume and sampled precision, then pulls wide with automatic sharding past Prospeo's 24k result ceiling.
  • Scale Qualification & Live Verification: Scores companies with a cheap OpenAI model against a tuned ICP prompt, re-verifies every qualified company's live website to remove dead or parked domains, and produces a client-facing HTML transparency report plus a verified-email contact ceiling.
  • Use Case: A client sells to multi-site medical groups but keyword searches return only 300 companies. Provide 10 known-fit domains, and the pipeline fingerprints them, generates thousands of lookalike candidates, mines discriminative filters, and delivers a verified qualified list with a TAM report.

Quick Start

Expand my list starting from these 10 seed company domains into a full qualified TAM for multi-site medical groups in the US, and generate the transparency report.

Frequently Asked Questions about list-expander

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

FAQPage Schema
How do I expand a small B2B lead list into a full TAM?

Start with about 10 seed domains known to fit your ICP, fingerprint how databases tag them, generate lookalike candidates via Prospeo, Exa, and Parallel.ai, then mine and score filters before pulling wide and AI-qualifying every company. The pipeline scripts handle each phase in order.

How to find lookalike companies with the Prospeo API?

Use the company_lookalike filter with a single seed domain per call, for example {"domain": "x.com"}, optionally combined with location or headcount filters. Arrays of domains return a 400 error, so the script makes one call per seed and paginates the results.

What API keys does the list expansion pipeline require?

PROSPEO_API_KEY and OPENAI_API_KEY are required for company search and AI qualification. EXA_API_KEY and PARALLEL_AI_API_KEY are optional lookalike lanes that are skipped with a log line when unset, so the run still completes on Prospeo alone.

Why does my Prospeo pull hit rate limits or truncate results?

The Prospeo account limit is global at roughly 150 requests per minute, so the scripts pace all processes through a shared slot file with a 450ms floor and a shared 45-second penalty on 429s. Filters exceeding about 24k results are auto-sharded by state and headcount band so the tail is never truncated.

How do I remove dead companies from a qualified lead list?

Run the verify-website script after AI qualification. It fetches every qualified domain's homepage live, classifies dead, suspended-parked, or live sites, and re-judges live sites against the ICP prompt, since stale database descriptions often qualify companies whose websites no longer exist.