devag7/linkedin-mcp

Overview

An MCP server that exposes LinkedIn profiles, people, job, and company search, feed, and messaging as structured JSON via 22 tools. Uses a stealth-browser engine to interact with LinkedIn's API instead of DOM scraping, with a built-in safety layer for gated writes like connect, message, post, react, and comment. Works with Claude, Cursor, or any MCP client.

Best for

Best for
Developers building AI assistants that need structured LinkedIn data and controlled social actions

Use cases

• Search and retrieve LinkedIn profiles and company data as structured JSON
• Automate job searches and feed monitoring through MCP clients
• Send connection requests or messages programmatically with safety controls

Notes

An MCP server that exposes LinkedIn profiles, people, job, and company search, feed, and messaging as structured JSON via 22 tools. Uses a stealth-browser engine to interact with LinkedIn’s API instead of DOM scraping, with a built-in safety layer for gated writes like connect, message, post, react, and comment. Works with Claude, Cursor, or any MCP client.

2 stars on GitHub. Last updated 2026-06-17. Licensed MIT.

Use cases

• Search and retrieve LinkedIn profiles and company data as structured JSON
• Automate job searches and feed monitoring through MCP clients
• Send connection requests or messages programmatically with safety controls

Pros

• Provides 22 tools covering read and write LinkedIn operations
• Uses API-level access instead of fragile DOM scraping
• Includes a safety layer for gated actions like messaging and posting

Cons

• Only 2 stars on GitHub, indicating limited community adoption or testing
• Requires a stealth-browser engine, adding complexity and potential maintenance overhead
• Gated writes may still trigger LinkedIn rate limits or detection

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.