Discover and explore top open-source AI tools and projects—updated daily.
LinkedIn data access for AI assistants
Top 73.8% on SourcePulse
This project provides a server for AI assistants, like Claude, to interact with LinkedIn data. It enables scraping profiles and companies, retrieving recommended jobs, and performing job searches, all accessible via a Docker container or local setup.
How It Works
The server leverages a headless Chrome browser instance, controlled via ChromeDriver, to navigate and scrape LinkedIn. It utilizes a session cookie (li_at
) for authentication, which needs to be periodically refreshed. The project is built upon the FastMCP framework and incorporates LinkedIn Scraper by joeyism.
Quick Start & Requirements
docker run -e LINKEDIN_COOKIE="li_at=YOUR_COOKIE_VALUE" stickerdaniel/linkedin-mcp-server:latest
uvx --from git+https://github.com/stickerdaniel/linkedin-mcp-server linkedin-mcp-server --cookie "li_at=YOUR_COOKIE_VALUE"
li_at
). The cookie expires within 30 days.Highlighted Details
--get-cookie
flag for automated cookie extraction, though manual extraction via Chrome DevTools is also detailed.Maintenance & Community
The project states all tools are functional and actively maintained as of July 2025. Users are encouraged to report issues on GitHub.
Licensing & Compatibility
Licensed under Apache 2.0. Use is subject to LinkedIn's Terms of Service, and web scraping may violate these terms. The tool is intended for personal use only.
Limitations & Caveats
LinkedIn's anti-scraping measures, including CAPTCHAs and mobile app confirmations, can disrupt automated cookie generation or login. Multiple active sessions with the same cookie can lead to invalidation. ChromeDriver version must match the installed Chrome browser.
1 month ago
Inactive