Browse
→ Communication
→ LinkedIn MCP
LinkedIn MCP
Python MCP server for LinkedIn-oriented workflows, built for Google Antigravity and compatible with Claude Code. It registers globally, exposes diagnostic and LinkedIn automation tools, uses a captured local browser session for search, comments, and connection requests, optionally uses a LinkedIn OAuth token for publishing, and stores job-scraping data in SQLite under ~/.config/linkedin-mcp. Public GitHub repository created September 3, 2026.
MCP unverified
Integration
| Transport | stdio |
| Auth | oauth |
| Endpoint | ./install.sh --claude |
| Install | git clone https://github.com/DeepLearner7/linkedin-mcp.git && cd linkedin-mcp && ./install.sh --claude |
Use Cases
| 01 | Expose LinkedIn search, comment, connection-request, and diagnostic workflows to an MCP client after local login |
| 02 | Run daily job scraping and recruiter-feed staging into a local SQLite database for agent review |
| 03 | Optionally publish LinkedIn posts through an OAuth access token while keeping credentials in local config |
Tags
linkedin jobs browser-session oauth sqlite recruiting automation python
Machine-readable: /api/servers.json
· JSON-LD schema embedded in <head>