intelligent-staffing-systems/mcp-linkedin

mcp-linkedin

MCP server for managing LinkedIn posts - create, update, and manage posts to share expertise and GitHub projects.

Overview

This MCP (Model Context Protocol) server enables programmatic management of LinkedIn posts. The primary use case is to help share technical expertise, showcase GitHub projects, and engage with a professional audience through LinkedIn's platform.

User Story

As a technical professional, I want to:

• Create LinkedIn posts that link to my GitHub projects
• Update existing posts to reflect project changes
• Manage my LinkedIn content programmatically
• Share my expertise and educational content with my professional network

This tool is designed to help educate and demonstrate expertise to a LinkedIn audience through consistent, high-quality content sharing.

Status

✅ MVP Complete - All core tools implemented and tested!

Milestones:

• ✅ LinkedIn Developer App with OAuth 2.0
• ✅ 7 MCP tools implemented (create, get, delete posts + auth)
• ✅ 24 automated tests passing
• ✅ Type-safe with JSDoc + Zod validation
• ✅ MCP server with stdio transport
• ✅ Ready for Claude Desktop integration

Features

✅ Implemented (MVP)

• linkedin_create_post - Create simple text posts with hashtags and mentions
• linkedin_create_post_with_link - Create posts with link previews (GitHub, blogs, etc.)
• linkedin_get_my_posts - Retrieve your recent posts with pagination
• linkedin_delete_post - Delete posts by URN
• linkedin_get_auth_url - Generate OAuth authorization URL
• linkedin_exchange_code - Exchange auth code for access token
• linkedin_get_user_info - Get your LinkedIn profile information

🚧 Planned (Next Phase)

• Post scheduling - Schedule posts for future publication (HIGH PRIORITY)
• Draft management - Save and manage draft posts locally
• Image uploads - Create posts with images
• Post updates - Edit existing posts
• Analytics - View post engagement metrics (if API supports)

LinkedIn API

This MCP will utilize LinkedIn's REST API:

• Base URL: https://api.linkedin.com/rest/posts
• Authentication: OAuth 2.0
• Required permissions: w_member_social scope
• API Documentation: LinkedIn Posts API

Development Philosophy

1. Start with clear user stories based on LinkedIn API capabilities
2. Refine requirements through discussion and iteration
3. Build incrementally with well-defined issues and branches
4. Test thoroughly with real-world use cases

Getting Started

Prerequisites

1. Create a LinkedIn Developer App at https://www.linkedin.com/developers/apps
2. Add these products to your app:
   • "Sign In with LinkedIn using OpenID Connect"
   • "Share on LinkedIn"
3. Configure OAuth redirect URI (e.g., https://localhost:3000/callback)

Setup

1. Clone the repository:

git clone https://github.com/intelligent-staffing-systems/mcp-linkedin.git
cd mcp-linkedin

2. Install dependencies:

npm install

3. Create .env file (copy from .env.example and fill in your values):

cp .env.example .env

4. Run tests:

npm test  # All 24 tests should pass

5. Setup with Claude Desktop:

See for detailed Claude Desktop configuration instructions.

Environment Variables

See .env.example for required configuration:

• LINKEDIN_CLIENT_ID - Your app's client ID
• LINKEDIN_CLIENT_SECRET - Your app's client secret
• LINKEDIN_REDIRECT_URI - OAuth callback URL
• LINKEDIN_PERSON_ID - Your LinkedIn person URN (from userinfo endpoint)
• LINKEDIN_ACCESS_TOKEN - OAuth access token (60-day expiry)

Documentation

• - Claude Desktop configuration and usage guide
• - Complete user stories and feature roadmap
• - Comprehensive API documentation
• - JSDoc type definitions
• - Zod validation schemas

Testing

npm test              # Run all tests
npm run test:watch    # Watch mode
npm run test:coverage # Coverage report

All 24 tests passing with 100% code coverage for core tools!

Contributing

Lucas Draney (@ldraney)

License

MIT