Linkedin-spider

You can deploy and run the Linkedin-spider MCP Server to automate data extraction from LinkedIn through a controlled, transport-based interface. This server lets you perform profile searches, post searches, profile scrapes, and company lookups, while managing sessions and avoiding detection when used responsibly within rate limits.

How to use

Set up a client that speaks the MCP protocol to the Linkedin-spider MCP Server. You can run the server locally and connect from your preferred MCP client or orchestration tool. Through the MCP interface, you can search for profiles and posts, fetch detailed profile information, scrape company data, and manage connections and conversations. Use the available transports to fit your environment and integrate with your existing automation.

How to install

# Prerequisites
- Python installed on your system (for MCP tooling and environment compatibility)
- Access to a shell/terminal

# Install the MCP server tooling (stdio transport usage shown as the primary example)
# Install via npm/pip/uvx depending on your setup is not assumed here; this example uses the provided MCP server command directly

# Start the MCP server using the stdio transport (local execution)
linkedin-spider-mcp serve stdio --email your_email@example.com --password your_password

# Alternatively, run via the development helper (uv)
uv run linkedin-spider-mcp serve stdio --email your_email@example.com --password your_password

Additional sections

Configuration and startup details below are shown as explicit command examples and environment variables you may use to customize the MCP server. You can tailor these values to your security and deployment requirements.

MCP server configuration and environment

Environment variables you may set for the MCP server, as illustrated in the startup examples, include authentication and transport controls. Use these as a basis for your deployment script or container environment.

Example usage patterns

• Start the MCP server with email/password authentication and the stdio transport for local development and scripting tasks.
• Start the MCP server with HTTP transport when you need remote access on a specific port and host.
• Use uv to run the MCP server in a development workflow. This helps with quick iterations and testing of MCP commands.

Security and responsible use

Respect rate limits and LinkedIn’s terms of service. Use the MCP server to collect data you are authorized to access, and avoid overburdening targets or spamming. Keep credentials secure and rotate them as needed.

Tools and capabilities (high level)

The server exposes capabilities to search profiles, search posts, scrape individual profiles, fetch company data, and manage connection data and conversations. Each capability is available through the MCP interface with dedicated commands and transports.

Available tools

search_profiles

Search for LinkedIn profiles using filters such as location, current company, and position. Returns a list of matched profiles with key details.

search_posts

Search LinkedIn posts by keywords and retrieve posts with metadata and links.

scrape_profile

Scrape a single LinkedIn profile to extract detailed information including experience, education, and skills.

scrape_company

Scrape company information from a LinkedIn company page, including industry, size, and location.

connections

Retrieve incoming and outgoing connection requests and manage connections data.

conversations

Get conversations list and detailed conversation history for messaging interactions.