# LinkFetch: LinkedIn MCP [Health: Active]

**Category:** 🌐 Social Media  
**Repository:** https://github.com/ffucucuoglu/linkedin-mcp-server  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/linkfetch-linkedin-mcp

## Description
LinkedIn jobs, companies and people for AI agents, plus messages and invites via your own account.

## Tools
Capabilities this server exposes over MCP:

- **search_jobs**
- **get_job**
- **search_companies**
- **get_company**
- **ai_score**
- **ai_classify**
- **ai_extract**
- **ai_write**
- **list_leads**
- **save_lead**
- **list_lead_lists**

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "linkfetch-linkedin-mcp": {
    "command": "npx",
    "args": ["-y","linkfetch-mcp"],
    "env": {
      "LINKFETCH_API_KEY": "",
      "LINKFETCH_MCP_URL": ""
    }
  }
}
```

**Requires environment variables:** `LINKFETCH_API_KEY`, `LINKFETCH_MCP_URL` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What LinkFetch: LinkedIn MCP MCP server does

LinkFetch: LinkedIn MCP MCP server exposes LinkedIn-oriented research and lead-management functions to MCP-compatible AI clients. Its listed tools cover job search and retrieval, company search and retrieval, AI processing, and lead lists:

- `search_jobs` and `get_job` find and inspect job records.
- `search_companies` and `get_company` find and inspect companies.
- `ai_score`, `ai_classify`, `ai_extract`, and `ai_write` process up to 25 jobs or companies at a time.
- `list_leads`, `save_lead`, and `list_lead_lists` work with saved leads.

The README also describes account-based LinkedIn operations for connected accounts, including people and company searches, profiles, employees, posts, connections, invitations, conversations, messages, comments, reactions, and follows. The available account actions and their limits are provided by the hosted service.

## How it works

The npm package is a small stdio bridge rather than a browser automation process. It opens an MCP connection to `https://linkfetch.io/mcp` and forwards tool-list and tool-call requests. Each request uses a LinkFetch API key, and the hosted endpoint supplies the current tools and pricing.

Clients that support remote MCP over Streamable HTTP can bypass the local package and connect to the same endpoint directly. The remote connection uses an `Authorization: Bearer` header. Claude can also use a custom connector URL containing the key.

Tool calls consume credits, with the flat cost stated in each tool description. Connected LinkedIn accounts are subject to daily and weekly limits enforced by LinkFetch. Errors such as rejected keys, insufficient credits, missing account connections, or reached limits are returned as tool results.

## Setup and configuration

To run LinkFetch: LinkedIn MCP MCP server locally, create an `lf_...` API key in the LinkFetch MCP dashboard and set `LINKFETCH_API_KEY`. Node.js 18 or later is required. Standard client configuration runs `npx -y linkfetch-mcp`; the README provides configurations for Claude Desktop, Claude Code, Cursor, Windsurf, Gemini CLI, Cline, VS Code, OpenAI Codex CLI, and Zed.

`LINKFETCH_MCP_URL` is optional and changes the hosted MCP endpoint from its default. Custom agents can spawn the process over stdio, including agents built with the OpenAI Agents SDK or LangChain MCP adapters. The repository also documents local development with `npm install`, `npm run build`, and `node dist/index.js`.

## Limitations and notes

LinkFetch: LinkedIn MCP MCP server depends on the hosted LinkFetch service, an active API key, available credits, and any required LinkedIn account connection. It does not run a browser on the local machine. The README states that older `sk_live_` keys no longer work after LinkFetch 2; use a newly created `lf_...` key instead. Connected-account operations remain subject to LinkFetch’s daily and weekly limits.

_Full upstream README: https://allmcps.com/mcp/linkfetch-linkedin-mcp/readme_