People Search

Overview

Find LinkedIn profiles of decision makers at target companies using Extruct’s company_people_finder. Takes a company table, adds a people finder column, and produces a linked child people table with names, roles, and LinkedIn URLs. No external API credits — uses Extruct’s index only.

When to Use

Trigger this skill when you need to:

Find LinkedIn profiles for target companies
Identify decision makers and contacts
Build a people list for email enrichment
Discover who to contact at each company

Trigger phrases: “find linkedin”, “find people”, “find contacts”, “find decision makers”, “people search”, “linkedin search”, “who to contact”, “find profiles”

list-segmentation → people-search → email-search → email-generation → email-response-simulation → campaign-sending

After you know WHICH companies to target, this skill finds WHO to contact. The next step (email-search) gets their verified emails and phones.

Environment Setup

EXTRUCT_API_TOKEN

string

required

Extruct API token for authentication. Set via:

export EXTRUCT_API_TOKEN=<your-token>

Base URL: https://api.extruct.ai/v1

Inputs

table_id

string

required

Extruct table ID from list-building or list-enrichment containing company data

target_roles

array

required

List of broad role descriptions to search for. Use 3-7 roles for best results.Examples:

["VP Sales", "Head of Sales", "Revenue Operations"]
["CTO", "VP Engineering", "Head of Data"]

max_results

number

default:"5"

Maximum number of people to find per company. Recommended: 3-5 (more = slower + noisier)

Workflow

Verify API Reference

The skill automatically:

Reads local reference: references/api_reference.md
Fetches live docs: https://www.extruct.ai/docs
Compares endpoints, params, and response fields
Updates local reference if discrepancies found
Flags changes to you before proceeding

Confirm the Table

Fetches table metadata via GET /tables/{table_id} and shows:

Table name
Table kind
Row count

Define Roles

You’ll be asked: “Who are we trying to reach at these companies?”The skill expands your input into a role list using broad role descriptions, not exact titles.

Role list guidelines:

3-7 roles is the sweet spot
Use broad terms: “VP Sales” not “Vice President of Enterprise Sales”
Cover the decision maker + influencer + champion if possible
Include department variations

Role Expansion Examples

You say: "Sales and ops leaders"
→ Roles: ["VP Sales", "Head of Sales", "Sales Operations", "RevOps", "Business Development", "CRO"]

You say: "People who buy data tools"
→ Roles: ["Revenue Operations", "Sales Operations", "Data Operations", "Business Intelligence", "CRM Admin"]

You say: "Innovation and strategy at corporates"
→ Roles: ["Head of Innovation", "VP Strategy", "Corporate Development", "Ventures", "Technology Scouting"]

The skill can also read role targets from your ICP in the context file:

claude-code-gtm/context/{company}_context.md → ## ICP → Roles column

Add company_people_finder Column

Creates a company_people_finder column via POST /tables/{table_id}/columns.Column configuration:

{
  "kind": "company_people_finder",
  "name": "Key Decision Makers",
  "key": "decision_makers",
  "value": {
    "roles": ["VP Sales", "Head of Sales", "Revenue Operations"],
    "provider": "research_pro",
    "max_results": 5
  }
}

Parameter	Description	Recommended
`roles`	Broad role descriptions to search for	3-7 roles
`provider`	Search provider	`research_pro` (always)
`max_results`	Max people per company	3-5

Trigger Enrichment Run

Runs the new column via POST /tables/{table_id}/run scoped to the new column ID.Reports: run ID and cells queued.

Monitor and Discover Child Table

The company_people_finder column auto-creates a child people table.The skill fetches parent table metadata and looks for child_relationships with relationship_type: "company_people" to get the people table ID.

Auto-created people table columns:

Column	Key	Kind	Description
Person Input	`input`	input	Raw person context string
Full Name	`full_name`	agent	Parsed full name
Role	`role`	agent	Current role/title
Profile URL	`profile_url`	agent (url)	LinkedIn URL

Optionally Add LinkedIn Data Column

For richer profile data (experience, education, skills), add a linkedin agent column to the people table:

{
  "kind": "agent",
  "key": "linkedin_data",
  "name": "LinkedIn Data",
  "value": {
    "agent_type": "linkedin",
    "prompt": "{profile_url}"
  }
}

This is optional — skip if you only need name + role + LinkedIn URL for the email-search step.

Fetch and Review People Data

Fetches data from the people table via GET /tables/{people_table_id}/data.Extracts: full_name, role, profile_url, and parent_row_id from each row.Summary presented:

Total companies in table: N
People found: N
Avg people per company: N
Companies with 0 people: N

Sample of 10 people shown for spot-checking.Review questions:

“Does the role mix look right?”
“Any roles missing? Want to add more?”
“Ready to proceed to email-search for verified emails?”

Export for email-search

Saves the people list as CSV at:

claude-code-gtm/csv/input/{campaign}/people_linkedin.csv

Columns: full_name, role, profile_url, parent_row_idThe email-search skill takes this CSV (or reads the people table directly) and enriches with verified emails.

Output

Output	Format	Location
People table	Extruct child table	Auto-created, linked to parent
People CSV	CSV with name, role, LinkedIn URL	`claude-code-gtm/csv/input/{campaign}/people_linkedin.csv`

Key Table IDs to Pass Forward

After this skill completes, pass these to email-search:

People table ID — for direct API access to people data
Parent table ID — for cross-referencing company data
Campaign slug — for file paths

API Reference

Create People Finder Column

POST /tables/{table_id}/columns
Authorization: Bearer {EXTRUCT_API_TOKEN}
Content-Type: application/json

Request:

{
  "kind": "company_people_finder",
  "name": "Key Decision Makers",
  "key": "decision_makers",
  "value": {
    "roles": ["VP Sales", "Head of Sales"],
    "provider": "research_pro",
    "max_results": 5
  }
}

Trigger Column Run

POST /tables/{table_id}/run
Authorization: Bearer {EXTRUCT_API_TOKEN}
Content-Type: application/json

Request:

{
  "column_ids": ["column-uuid"]
}

Fetch People Table Data

GET /tables/{people_table_id}/data
Authorization: Bearer {EXTRUCT_API_TOKEN}

Response:

{
  "rows": [
    {
      "id": "row-uuid",
      "parent_row_id": "parent-row-uuid",
      "cells": {
        "full_name": "Jane Doe",
        "role": "VP of Sales",
        "profile_url": "https://linkedin.com/in/janedoe"
      }
    }
  ]
}

Next Steps

After people search completes:

Proceed to email-search to get verified emails and phones
Or manually review the people list and refine role filters
Or add more companies to the parent table and re-run

Get Started

Core Concepts

Skills Reference

Guides

API Integration

Overview

When to Use

Environment Setup

Inputs

Workflow

Output

Key Table IDs to Pass Forward

API Reference

Create People Finder Column

Trigger Column Run

Fetch People Table Data

Next Steps

Build docs developers (and LLMs) love

Get Started

Core Concepts

Skills Reference

Guides

API Integration

​Overview

​When to Use

​Related Skills

​Environment Setup

​Inputs

​Workflow

​Output

​Key Table IDs to Pass Forward

​API Reference

​Create People Finder Column

​Trigger Column Run

​Fetch People Table Data

​Next Steps

Build docs developers (and LLMs) love

Overview

When to Use

Related Skills

Environment Setup

Inputs

Workflow

Output

Key Table IDs to Pass Forward

API Reference

Create People Finder Column

Trigger Column Run

Fetch People Table Data

Next Steps