OutX MCP Server

MCP server for the OutX LinkedIn social listening and data API. Connect AI agents like Claude, Cursor, and Claude Code to LinkedIn — create watchlists, search posts, fetch profiles, and engage, all through natural language.

AI Agent ←→ MCP Protocol ←→ OutX MCP Server ←→ OutX API ←→ LinkedIn

What You Can Do

• Monitor LinkedIn — Track posts by keywords, people, or companies with watchlists
• Search & filter posts — 15+ filters: keyword, date, seniority, trending, language
• Fetch profiles & companies — Get full LinkedIn profiles, company pages, and connections
• Search LinkedIn — Find people by title, company, location, industry, and more
• Message connections — Send direct messages to 1st-degree connections
• Engage — Like and comment on posts through your LinkedIn account
• All via natural language — Just tell your AI agent what you want

Quick Start

1. Get an API key at mentions.outx.ai/api-doc

2. Add to your AI tool (pick one):

{
  "mcpServers": {
    "outx": {
      "command": "npx",
      "args": ["-y", "outx-mcp-server"],
      "env": {
        "OUTX_API_KEY": "your-api-key"
      }
    }
  }
}

{
  "mcpServers": {
    "outx": {
      "command": "npx",
      "args": ["-y", "outx-mcp-server"],
      "env": {
        "OUTX_API_KEY": "your-api-key"
      }
    }
  }
}

{
  "mcpServers": {
    "outx": {
      "command": "npx",
      "args": ["-y", "outx-mcp-server"],
      "env": {
        "OUTX_API_KEY": "your-api-key"
      }
    }
  }
}

{
  "mcpServers": {
    "outx": {
      "command": "npx",
      "args": ["-y", "outx-mcp-server"],
      "env": {
        "OUTX_API_KEY": "your-api-key"
      }
    }
  }
}

3. Start using it:

"Create a watchlist to monitor LinkedIn posts about AI startups raising Series A"

Available Tools (26)

Watchlist Management

Pause and resume any watchlist with disable: true or disable: false on its update tool. Configure Slack alerts with slack_webhook_url.

Posts & Engagement (operates on posts your watchlists captured)

LinkedIn Data API (direct LinkedIn actions, no watchlist required)

Async tools, the MCP server polls LinkedIn for ~150s and returns the result directly. You almost never need get_task_status because of this.

Note: LinkedIn Data tools require the OutX Chrome extension to be active within the last 48 hours.

Example Prompts

Once connected, try these with your AI agent:

Social listening:

• "Create a watchlist to monitor LinkedIn posts about AI startups raising Series A"
• "Show me trending posts from my watchlists this week"
• "What are VPs and Directors posting about machine learning?"
• "Find posts mentioning our competitor from the last 7 days"

Profile & company research:

• "Fetch the LinkedIn profile for williamhgates"
• "Get recent posts from this prospect's profile"
• "Get company info and recent posts for OpenAI"
• "Search for VPs of Engineering at Google in the US"
• "Show me my most recently added connections"

Engagement & messaging:

• "Like the top 3 most engaging posts about product management"
• "Comment on this post with a thoughtful response"
• "Send a message to this connection thanking them for connecting"

Workflows:

• "Set up a watchlist for 'looking for a CRM' and check it every 6 hours"
• "List all my watchlists and show me posts from the sales signals one"

Prerequisites

• Node.js 18+ — Required for the MCP server runtime
• OutX account — Sign up free and get your API key
• OutX Chrome extension — Required for LinkedIn Data API tools (profile fetching, direct engagement). Install here

How It Works

The OutX API has two surfaces:

1. Watchlists & Engagement API (sync) — Create watchlists to monitor LinkedIn by keywords, people, or companies. OutX scans on your schedule and stores posts. Query and engage via API.

2. LinkedIn Data API (async) — Direct LinkedIn proxy. Fetch profiles, get posts, like, comment. All endpoints return a task ID; the MCP server polls automatically until the result is ready.

The MCP server wraps both APIs into 28 tools that any MCP-compatible AI agent can call.

Development

git clone https://github.com/outxai/outx-mcp-server.git
cd outx-mcp-server
npm install
npm run build
OUTX_API_KEY=your-key npm start

For development with auto-reload:

OUTX_API_KEY=your-key npm run dev

Documentation

• MCP Integration Guide — Detailed setup and configuration
• API Reference — Full REST API documentation
• API Quick Start — Create watchlists and retrieve posts
• LinkedIn Data API Quick Start — Fetch profiles in 2 minutes
• Authentication Guide — API keys and OTP auth
• Rate Limits — Usage limits and best practices
• API Use Cases — Sales signals, competitor monitoring, hiring alerts
• Chrome Extension — Required for LinkedIn Data tools
• LinkedIn Safety Guide — Rate limiting and account protection

Integrations

• Python SDK — Drop-in Python wrapper for the OutX API
• LangChain Integration — Use OutX as a LangChain tool for AI agents
• n8n Integration — No-code LinkedIn automation workflows

Guides & Tutorials

• How to Connect AI Agents to LinkedIn — LangChain, Claude, GPT, and MCP examples
• LinkedIn API Guide — Comprehensive API overview
• LinkedIn API Alternatives 2026 — OutX vs Proxycurl vs PhantomBuster vs Unipile
• How to Build a Sales Signal Pipeline — Detect buying intent with the OutX API
• LinkedIn Competitor Monitoring Guide — Track competitor activity via API
• Social Listening on LinkedIn — Complete guide to LinkedIn monitoring
• How to Automate LinkedIn Likes & Comments — Engagement automation best practices
• LinkedIn Automation Safety Guide — Keep your account safe

Links

• OutX Platform — Sign up and get started
• Get API Key — Generate your API key
• OutX for AI Agents — All AI integration options (MCP, skill file, llms.txt)
• Pricing — Free, Pro, and Enterprise plans
• FAQ — Common questions answered

Partnerships

Interested in a collaboration or affiliate partnership? Drop us a message at support@outx.ai.

License

MIT