LinkedIn

This MCP server lets you query live LinkedIn data through a simple, read-only MCP interface. By wrapping the CData JDBC Driver for LinkedIn, you can ask natural language questions and retrieve up-to-date LinkedIn information without writing SQL. It’s designed for local querying by AI clients that need fast access to LinkedIn data sources.

How to use

You connect an MCP client to the LinkedIn MCP server and use built-in tools to explore and read data. Start by ensuring the server is running on the same machine as your MCP client, then issue questions that map to tables and columns in the LinkedIn data model. You can retrieve the list of available tables, inspect their columns, and run read-only queries that fetch live results from LinkedIn via the JDBC driver.

How to install

Prerequisites: Java and Maven installed on your system. You will also need the CData JDBC Driver for LinkedIn to connect to LinkedIn data.

Step 1: Clone the project and enter the directory.

git clone https://github.com/cdatasoftware/linkedin-mcp-server-by-cdata.git
cd linkedin-mcp-server-by-cdata

Step 2: Build the server JAR with dependencies.

mvn clean install

Step 3: Download and install the CData JDBC Driver for LinkedIn from the vendor site. Follow the driver’s installer prompts and license steps.

Step 4: License the CData JDBC Driver.

java -jar cdata.jdbc.linkedin.jar --license

Step 5: Configure the connection to LinkedIn (via the Driver’s Connection String utility). Save the resulting connection string for later use.

Step 6: Create a .prp file (for example linkedin.prp) with your JDBC connection details. Include a Prefix, ServerName, ServerVersion, DriverPath, DriverClass, JdbcUrl, and Tables as shown.

Prefix=linkedin
ServerName=CDataLinkedIn
ServerVersion=1.0
DriverPath=PATH\TO\cdata.jdbc.linkedin.jar
DriverClass=cdata.jdbc.linkedin.LinkedInDriver
JdbcUrl=jdbc:linkedin:InitiateOAuth=GETANDREFRESH;
Tables=

Using the server with Claude Desktop (example client)

To connect Claude Desktop or another MCP client, add a server entry that points to the JAR and .prp file you created. Place the configuration in the client’s MCP config file under mcpServers.

Windows example entry (adjust paths to your environment):

{
  "mcpServers": {
    "linkedin": {
      "command": "PATH\\TO\\java.exe",
      "args": [
        "-jar",
        "PATH\\TO\\CDataMCP-jar-with-dependencies.jar",
        "PATH\\TO\\linkedin.prp"
      ]
    }
  }
}

Linux/macOS example entry (adjust paths to your environment):

{
  "mcpServers": {
    "linkedin": {
      "command": "/PATH/TO/java",
      "args": [
        "-jar",
        "/PATH/TO/CDataMCP-jar-with-dependencies.jar",
        "/PATH/TO/linkedin.prp"
      ]
    }
  }
}

Running the server

Start the MCP server on its own machine using the prepared JAR and the LinkedIn PRP file. The server runs in stdio mode and is intended for use with clients that operate on the same host.

java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/Salesforce.prp

Usage details

Once configured, the AI client can read LinkedIn data and invoke the built-in tools to query data without needing explicit SQL. Typical questions include asking for correlations, counts of items, or calendar-type data derived from LinkedIn data sources.

Available tools

linkedin_get_tables

Retrieves a list of tables available in the LinkedIn data source. The output is in CSV format with a header row.

linkedin_get_columns

Retrieves a list of columns for a specific LinkedIn table. The output is in CSV format with a header row.

linkedin_run_query

Executes a SQL SELECT query against the LinkedIn data source and returns results.