Skip to main content
Guide1 min read

LinkedIn MCP server: what works and what to watch (2026)

Searches for a LinkedIn MCP are rising. Here's the honest landscape — community servers, the official API path, and the big ToS caveats.

"LinkedIn MCP" is a fast-rising search, so here's the honest picture: LinkedIn has no broad official MCP server, and the community options that scrape your account live in a grey area with real account-risk. This guide covers what exists and how to use it carefully.

Caveat first: LinkedIn's terms restrict automated access and scraping. Community servers that log in as you can get your account flagged or banned. Proceed knowingly, and prefer official APIs where your use case allows.

The landscape

  • Community scraper servers (e.g. self-hosted servers that drive a logged-in session) can read profiles, search and pull post data. Powerful, but they use unofficial access — that's the ToS and ban risk.
  • Official LinkedIn APIs (Marketing, Share, Sign In) are sanctioned but narrow and require an approved developer app — better for posting/marketing use cases than for broad scraping.
  • Bridges (Composio, Pipedream) expose LinkedIn actions through their managed integrations within LinkedIn's allowed scopes.

Typical community setup

Most run via Docker or uv and need your LinkedIn session/cookie:

{
  "mcpServers": {
    "linkedin": {
      "command": "uvx",
      "args": ["<community-linkedin-mcp-package>"],
      "env": { "LINKEDIN_COOKIE": "your_li_at_cookie" }
    }
  }
}

Swap in the exact package from the repo you vet. Restart the client; see MCP config file location.

Safer path

For posting and analytics, use an official LinkedIn developer app + a bridge like Composio that stays within sanctioned scopes — lower ban risk than a cookie-based scraper.

Use responsibly

Read your session cookie like a password, never share it, and understand the account risk before automating LinkedIn. Review any community server's code first — see how to vet an MCP server and MCP security best practices.

Going further

Browse the communication category or curated loadouts.

Loadout

Build your AI agent loadout

The directory of MCP servers and AI agents that actually work. Pick the right loadout for Slack, Postgres, GitHub, Figma and 20+ integrations — with install commands ready to paste into Claude Desktop, Cursor or your own stack.

© 2026 Loadout. Built on Angular 21 SSR.