linkedin-scraper_skill

This skill assists you scrape LinkedIn profiles and build lead lists by leveraging your Chrome session for authenticated data extraction.

Python

1.7k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill linkedin-scraper

_meta.json285 B
SKILL.md5.5 KB

Overview

This skill scrapes LinkedIn profiles and search results using your authenticated Chrome browser session so no API keys are required. It extracts structured contact and profile data and can insert results into a DuckDB workspace for lead management. Use it to build prospect lists, enrich contacts, or automate high-volume profile collection while following rate limits and stealth patterns.

How this skill works

The skill drives your Chrome profile via a browser relay: it opens LinkedIn search pages or profile URLs, takes DOM snapshots, and parses structured fields (name, headline, location, current title, company, experience, education, connections, about, profile URL). For bulk work it iterates search result cards, visits profile pages with randomized delays and natural scrolling, and optionally inserts records into a DuckDB table or returns JSON arrays. It watches for sign-in prompts, CAPTCHAs, and rate-limit signals and stops to alert the user when needed.

When to use it

Find founders, executives, or prospects on LinkedIn using keyword or company searches
Scrape a single LinkedIn profile to extract structured contact and role data
Bulk-scrape search results to build a prospect or outreach list
Populate or update a DuckDB leads/contacts table with LinkedIn-sourced records
Verify and enrich an existing contact list with LinkedIn profile details

Best practices

Always use an active, legitimate Chrome login for the target LinkedIn account
Respect the rate limits: 3–8s randomized delays, max ~80 profiles per session, 60s breaks after ~20 profiles
Enable natural scrolling and varied visit order to reduce detection risk
Stop immediately and notify if a CAPTCHA, security check, or sign-in page appears
Store scraped data locally (DuckDB) and follow privacy and legal guidelines

Example use cases

Search for Y Combinator founders and extract names, titles, companies, and profile URLs for outreach
Scrape a list of company employees from a company People page and load into DuckDB as leads
Enrich CRM records by visiting LinkedIn profile URLs and returning current title, company, and about text
Run nightly small-batch updates (under rate limits) to refresh prospect information before a campaign

FAQ

The skill stops immediately, alerts you, and recommends waiting at least 30 minutes and rechecking the Chrome session before resuming.

How is data delivered?

Default output is structured JSON per profile; optional integration can insert records into a DuckDB leads/contacts object.

How many profiles can I safely scrape per session?

Follow the rule of thumb: maximum ~80 profiles per session, pause 60s after every 20 profiles, and use randomized 3–8s delays between page loads.

What fields are extracted from a profile?

Name, headline, location, current title/company, experience, education, connections count, about text, and canonical LinkedIn URL.