Log inSubmit server
LinkedIn logo

LinkedIn

Read LinkedIn data and manage messages or connection requests through your logged-in browser session.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install LinkedIn

Pick your client, then add it in one click or paste the config.

STDIOPython
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "linkedin": {
      "command": "uvx",
      "args": [
        "linkedin-api-mcp"
      ]
    }
  }
}

Settings ā†’ Developer ā†’ Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

We couldnā€™t confirm it starts correctly

We ran the install command but it didn't respond within our test window ā€” this can mean a slow first-time install rather than a real problem.

uvx linkedin-api-mcp

No response to initialize.

This is an experimental automated check and can have false negatives ā€” missing environment variables, a slow cold install, etc. It doesnā€™t necessarily mean somethingā€™s wrong. Last checked 23d ago.

Overview

The LinkedIn MCP server exposes LinkedIn profile, company, job, post, and messaging operations through MCP. It drives a headless browser with your logged-in LinkedIn session cookie rather than using an API key or scraping API. Read tools cover discovery and inbox access, while message sending and connection requests are separate write actions. Use it when an agent needs LinkedIn data or account actions tied to your own session.

Use cases

  • Research member profiles and work histories
  • Search companies and inspect their details
  • Find jobs by keyword and location
  • Review inbox threads and conversation history
  • Send messages or connection invitations

Key features

  • Profile and company lookup
  • People, company, job, and post search
  • Inbox and conversation access
  • Message and connection-request actions
  • Browser-session authentication
  • Session and usage status checks

Tools12

What LinkedIn lets your agent do. Expand a tool for its parameters.

Self-reported~200 tokens of context

Security & health

What it can reach and how reliably it runs ā€” from our own checks, not the maintainerā€™s claims.

ActiveLast checked 2d ago

Recent health check succeeded.

12/14 checks healthy over the last 90d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 1d ago via OSV.dev Ā· linkedin-api-mcp (PyPI)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks ā€” not a rating.

GitHub stars4Stargazers on the source repository.
Last commit2mo agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window ā€” often a slow first install.
Tools exposed12Callable tools this server registers over MCP.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How LinkedIn works

What the LinkedIn MCP server does

The LinkedIn MCP server gives an MCP-compatible AI client access to LinkedIn through a browser session associated with your account. It can retrieve member profiles by URL or public identifier, inspect your own profile, search for people, and look up company pages. Company results include details such as about information, size, industry, and recent posts.

Job discovery is supported through keyword and optional location searches, with a separate tool for retrieving a complete job posting. The server can also search LinkedIn feed posts. For messaging, it lists recent inbox thread previews, retrieves the full history for a selected thread, and searches conversations by participant or keyword.

Two actions change account state: send_message sends a real LinkedIn message, and connect sends a real connection invitation. They are kept separate from the read tools so an agent can distinguish account changes from information retrieval.

How it works

The LinkedIn MCP server runs a headless Chromium browser using Patchright, an undetected Playwright fork. It uses the li_at session cookie from an existing LinkedIn login. This is browser-session access, not a LinkedIn API integration, credential-stuffing flow, or separate scraping API.

All operations use one browser session and run one at a time. Requests are paced according to limits configured by the user. The linkedin_status tool checks whether the session remains valid and reports current usage against those limits.

Setup and configuration

Install the server with uvx linkedin-api-mcp. Before using it, install the Chromium build required by the server with uvx --from linkedin-api-mcp patchright install chromium. Log in to LinkedIn in a regular browser, copy the value of the li_at cookie from the LinkedIn siteā€™s application storage, and provide it through uvx linkedin-api-mcp auth.

By default, authentication data is stored in the operating system keyring: Keychain on macOS, Credential Manager on Windows, or Secret Service on Linux. The server also supports passing the cookie as LINKEDIN_COOKIE, which can be useful in sandboxed environments or CI. Treat that value as equivalent to a login credential and protect any configuration file containing it.

Run uvx linkedin-api-mcp --test to confirm that LinkedIn accepts the session. The command prints redacted configuration details and does not print the cookie. Client configuration uses a stdio command with uvx and linkedin-api-mcp; documented configurations are available for Claude Code, Claude Desktop, Cursor, VS Code, Codex CLI, Zed, and Windsurf.

Tools and capabilities

The LinkedIn MCP server provides tools for:

  • Reading member profiles and the logged-in memberā€™s profile.
  • Searching people, companies, jobs, and posts.
  • Reading company details and full job postings.
  • Listing, searching, and opening message conversations.
  • Sending a message or connection invitation.
  • Checking session validity and usage status.

These capabilities suit research, recruiting workflows, job discovery, and inbox-oriented tasks where the agent must work with data visible to the logged-in account.

Limitations and notes

The session cookie grants access as the associated LinkedIn account. Anyone who obtains it may be able to read that accountā€™s messages and visible LinkedIn data, send messages, and send invitations. Do not place it in prompts, source control, screenshots, issues, or shared configuration.

This is an independent community project and is not affiliated with, authorized by, endorsed by, or sponsored by LinkedIn Corporation or Microsoft. The cookie-based approach also means that access depends on the browser session remaining valid and on what the account can see. Write tools perform real account actions, so clients should obtain appropriate confirmation before using them.

Read the full README View source on GitHub

Reviews

No reviews yet ā€” be the first to share how it worked for you.

Frequently Asked Questions about LinkedIn

LinkedIn is an MCP server that connects an AI client to a logged-in LinkedIn browser session through MCP. Its main tools read profiles, companies, jobs, posts, inboxes, and conversations, while send_message and connect perform separate write actions. It uses the accountā€™s session cookie rather than a LinkedIn API key.

Install it with `uvx linkedin-api-mcp`, then install the required Chromium build with `uvx --from linkedin-api-mcp patchright install chromium`. Configure the server as a stdio MCP command in your client, and authenticate with `uvx linkedin-api-mcp auth`.

It authenticates with your LinkedIn `li_at` session cookie. Store the cookie through the serverā€™s `auth` command for OS keyring storage, or provide it with `LINKEDIN_COOKIE` when needed for environments such as CI. The cookie should be protected like a password.

Yes, LinkedIn works with Cursor through a documented MCP install link or manual stdio configuration. The Cursor configuration runs `uvx` with `linkedin-api-mcp` as its argument, and the LinkedIn session cookie must be configured separately.

Related MCP servers

All in Social Media All alternatives
Add a badge to your READMEShow itā€™s listed ā€” live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewLinkedIn AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/linkedin?style=directory)](https://allmcps.com/mcp/linkedin)
HTML Embed
<a href="https://allmcps.com/mcp/linkedin"><img src="https://allmcps.com/api/badge/linkedin?style=directory" alt="LinkedIn on AllMCPs" /></a>