LinkedIn MCP Server
A MCP server to access the LinkedIn API through RapidAPI, based on the FastMCP framework.
Features
- Search for LinkedIn profiles
- Get detailed information about profiles
- Get recent posts from profiles
Requirements
- Python 3.12 or higher
- A RapidAPI API key for LinkedIn Data API
- UV (fast Python installer)
Installation
Install UV
If you don’t have UV installed, you can install it with the following command:
curl -LsSf https://astral.sh/uv/install.sh | sh
Install the MCP server
Clone this repository and navigate to the folder:
git clone https://github.com/fcojaviergon/linkedin-mcp.git cd linkedin-mcpInstall dependencies using UV:
uv pip install -e .Create a
.envfile with your API key:RAPIDAPI_KEY=your_rapidapi_key_here
Usage
Run the server directly
To run the server manually:
uv run mcp run server.py
Configure with Claude Desktop
Add this configuration to your MCP configuration file:
- For Claude Desktop:
/path/to/Claude/claude_desktop_config.json
{
"mcpServers": {
"LinkedIn MCP Server": {
"command": "/path/to/uv/.local/bin/uv",
"args": [
"run",
"--with",
"mcp[cli]",
"mcp",
"run",
"/path/to/projects/rapidapi-linkedin-mcp/server.py"
]
}
}
}
Make sure to replace /path/to/uv/.local/bin/uv with the actual path to your UV installation and /path/to/projects/rapidapi-linkedin-mcp/server.py with the full path to the server.py file.
Tools available
search_people
Search for LinkedIn profiles with various filters.
Parameters:
keywords(optional): Keywords for the searchstart(optional): Pagination start point (0, 10, 20, etc.)geo(optional): Location IDfirst_name(optional): First namelast_name(optional): Last namecompany(optional): Company nameschool_id(optional): School IDkeyword_school(optional): School keywordkeyword_title(optional): Title keyword
get_profile
Get detailed information about a LinkedIn profile.
Parameters:
username(required): LinkedIn username
get_profile_posts
Get recent posts from a LinkedIn profile.
Parameters:
username(required): LinkedIn usernamestart(optional): Pagination start point (0, 10, 20, etc.)pagination_token(optional): Pagination tokenposted_at(optional): Filter by post date
Prompts available
search_people_prompt: To search for profilesprofile_prompt: To get information about profilesprofile_posts_prompt: To get recent posts from profiles
Examples of usage from an agent
from mcp.client import MCPClient
# Connect to the server
client = MCPClient("http://localhost:8000")
# Search for profiles
result = await client.invoke_tool("search_people", {
"keywords": "developer",
"company": "Google"
})
# Get profile
profile = await client.invoke_tool("get_profile", {
"username": "satyanadella"
})
# Get recent posts
posts = await client.invoke_tool("get_profile_posts", {
"username": "billgates"
})
Troubleshooting
- If you receive an authentication error, verify that your API key in the
.envfile is correct - Make sure you have an active subscription to the LinkedIn endpoint in RapidAPI
- If you have issues with UV, you can use pip directly:
pip install -e .
LinkedIn API using RapidAPI
Project Details
- fcojaviergon/rapidapi-linkedin-mcp
- MIT License
- Last Updated: 5/22/2025
Recomended MCP Servers
An open-source MCP server for integrating Wazuh security data with LLMs (such as the Claude Desktop App). This...
An MCP interface into the uProc toolset
query table from some websites, support MCP
MCP for devcontainers
simple mcp server to wrap the local instance of grep.
Model Context Protocol server for DeepSeek's advanced language models
pubmed-mcp-smithery





