skills.rest
narrative-io
narrative-ioOFFICIAL
Explore all skills by this author →
WebsiteGitHub

create-lookalike

Automate look-alike audience creation by scoring non-seed candidates with Rosetta Stone mappings.

7|Updated May 18, 2026
Search Tags:#workflow#narrative#pipeline#audience#segmentation#lookalike#rosetta-stone
One-click install
npx skills add https://github.com/narrative-io/narrative-skills-marketplace --skill create-lookalike
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: create-lookalike
Source: https://github.com/narrative-io/narrative-skills-marketplace/tree/main/plugins/narrative-audience/skills/create-lookalike
Command: npx skills add https://github.com/narrative-io/narrative-skills-marketplace --skill create-lookalike

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires [], and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

Create accurate look-alike audiences from a seed and a candidate population by orchestrating a deterministic scoring pipeline that mirrors Lookalike Studio templates and enforces review gate.

Core Features & Use Cases

  • Build a structured look-alike workflow by expanding identities, scoring non-seed candidates, and presenting an auditable pipeline for approval.
  • Leverage Rosetta Stone mappings to align seed and population identities, apply feature-based scoring (categorical and continuous), and produce an output audience ready for connectors or Lookalike Studio edits.
  • Use cases include expanding customer bases, finding similar users, and validating audience quality before activation.

Quick Start

Provide the seed and population dataset identifiers, then choose an output mode (size or score) to generate the look-alike audience.

Frequently Asked Questions about create-lookalike

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

FAQPage Schema
How do I create a lookalike audience from a seed dataset?▼

To create a lookalike audience, provide seed and population dataset identifiers, then choose an output mode (size or score) to expand identities, score non-seed candidates, and produce a final deliverable pending human approval.

How does identity mapping work when building audience segmentation pipelines?▼

Identity mapping uses Rosetta Stone mappings to align seed and population identities, enabling feature-based scoring across categorical and continuous data to accurately expand your seed audience into a lookalike segment.

Can I re-edit a lookalike audience after it has been generated?▼

Yes, you can re-edit generated lookalike audiences by utilizing Lookalike Studio wizard-state encoding via the lookalike_state_tag.py script, which preserves the pipeline state for future modifications.

Do I need Narrative MCP to resolve datasets for lookalike modeling?▼

Yes, creating lookalike audiences requires Narrative MCP-backed dataset resolution and NQL validation to ensure accurate identity expansion and feature scoring before the pipeline gates on human approval.

What is the best way to validate audience quality before activation?▼

The best way to validate audience quality is to run the deterministic scoring pipeline, which applies Rosetta Stone mappings and feature-based scoring to produce an auditable pipeline for review before activation.

Related Skills

Explore More Data & Analytics Skills→
narrative-ionarrative-io
7

generate-rosetta-stone-mappings

Generate and validate Rosetta Stone mappings for Narrative datasets with NQL validation.

Official
Advanced
wattdatawattdata
5

audience-generate-list

Generate matched, expanded, lookalike, and scored audiences from owned lists.

Official
Advanced
narrative-ionarrative-io
7

write-nql

Draft natural-language questions into validated NQL queries for Narrative datasets.

Official
Advanced
cristian473cristian473

free-tool-strategy

Plan and evaluate free marketing tools for lead generation.

Community
Intermediate