| name | linkedin-profile-scraper |
| description | Scrapes and extracts data from LinkedIn profile pages. Use when user provides a linkedin.com/in/URL and wants profile data extracted - includes name, headline, about, experience, education from the profile. Requires authenticated session cookies stored in .config/linkedin-cookies.json. Outputs JSON with extracted profile text for stakeholder analysis or CRM ingestion. Do NOT use for unauthorized data collection - only with user's own authenticated LinkedIn session.
|
| compatibility | ["playwright"] |
LinkedIn Profile Scraper
Overview
This skill scrapes LinkedIn profiles using Playwright with authenticated cookies. It extracts:
- Name, headline, location
- About section
- Experience and education
- Connection count
Prerequisites
- Cookie file must exist at
.config/linkedin-cookies.json
- Playwright must be installed:
pip install playwright && playwright install chromium
Workflow
Step 1: Validate cookie file exists
Read .config/linkedin-cookies.json. If missing, error: "Cookie file not found. Export cookies from LinkedIn using a browser extension."
Step 2: Load cookies
Parse the JSON. Validate required fields: li_at (primary auth token).
Step 3: Build scraper script
Write /tmp/linkedin_scraper.py with:
- Load cookies from config path
- Launch headless Chromium with cookies
- Navigate to profile URL
- Extract text content from page
- Output JSON with
source, source_url, extracted_text
Step 4: Run scraper
Execute: python /tmp/linkedin_scraper.py
Step 5: Parse results
The script outputs JSON to /tmp/{profile_name}_ingest.json. Read and present extracted data to user.
Usage
User: "ingest https://www.linkedin.com/in/username"
→ Extract profile URL → Run workflow → Present extracted data
Output Format
{
"source": "linkedin",
"source_url": "https://www.linkedin.com/in/username",
"extracted_text": "Full profile text..."
}
Notes
- Cookies expire ~1 year. Re-export if 401 errors occur.
- LinkedIn rate-limits heavy usage. Space out requests.
- Headless mode may fail if LinkedIn updates anti-bot measures.
