Linkedin MCP Server — Hosted Api Connectors Integration

mcp-linkedin — LinkedIn MCP Server

Automate LinkedIn from any AI agent — manage profiles, publish posts, engage with company pages, search jobs, and send messages using the full LinkedIn REST API.

LinkedIn is the world's largest professional network with over 900 million members. This MCP server gives your AI agents complete access to the LinkedIn REST API: reading and updating profiles, publishing personal and company posts, searching jobs, managing conversations, liking and commenting on content, and tracking company followers.

Live endpoint: https://mcp.aerostack.dev/s/aerostack/mcp-linkedin

---

What You Can Do

• Automated content publishing — Schedule and post thought leadership content on behalf of team members or company pages without touching the LinkedIn UI
• Lead generation pipeline — Search for companies by keyword, pull their follower demographics, find decision-makers via connections, and send personalized outreach messages
• Competitor monitoring — Track a competitor's company page posts and engagement metrics on a cron schedule, then summarize weekly findings
• Job market intelligence — Search job listings for specific roles and locations, extract salary and requirements data, and build a structured talent pipeline
• Community engagement — Automatically like and comment on relevant industry posts based on keyword monitoring to grow organic reach

Available Tools

Tool	Description
get_my_profile	Get the authenticated user's profile (name, headline, summary, location)
get_profile_by_id	Get a LinkedIn member profile by URN or ID
get_connections	Get first-degree connections list (up to 500)
get_profile_views	Get who viewed your profile in the last 90 days
create_post	Create a text post (PUBLIC or CONNECTIONS visibility)
create_post_with_image	Create a post with an image from a URL
delete_post	Delete a post by URN
get_post	Get a specific post by URN
like_post	Like a LinkedIn post
comment_on_post	Add a comment to a post
get_company	Get company details (name, description, followers, industry)
get_company_posts	Get recent posts from a company page
create_company_post	Post on behalf of a company page (requires page admin)
get_company_followers	Get follower count and demographic summary
search_companies	Search for companies by name or keyword
search_jobs	Search LinkedIn jobs by keywords, location, or company
get_job	Get full job details by job ID
send_message	Send a direct message to a first-degree connection
get_conversations	Get list of message conversations
get_conversation_messages	Get messages in a specific conversation
_ping	Verify credentials by calling /v2/me

Configuration

Variable	Required	Description	How to Get
LINKEDIN_ACCESS_TOKEN	Yes	OAuth 2.0 User Access Token with appropriate scopes	LinkedIn Developer Portal → Your App → Auth → OAuth 2.0 Tools → Generate Token

Required OAuth scopes:

• r_liteprofile — Read basic profile information
• r_emailaddress — Read email address
• w_member_social — Write posts, likes, and comments
• rw_organization_admin — Read/write company page content (required for company tools)

Quick Start

Add to Aerostack Workspace

1. Go to aerostack.dev → Your Project → MCPs
2. Search for "LinkedIn" and click Add to Workspace
3. Add your LINKEDIN_ACCESS_TOKEN under Project → Secrets

Once added, every AI agent in your workspace can interact with LinkedIn professionally.

Example Prompts

"Post a LinkedIn update saying we just launched our new AI product, visibility PUBLIC"
"Get the last 5 posts from company ID 1234567 and summarize the engagement themes"
"Search for backend engineer jobs in San Francisco and return the top 10 with company names"
"Get my LinkedIn profile and tell me how to improve my headline for AI developer roles"

Direct API Call

curl -X POST https://mcp.aerostack.dev/s/aerostack/mcp-linkedin \
  -H 'Content-Type: application/json' \
  -H 'X-Mcp-Secret-LINKEDIN-ACCESS-TOKEN: your-access-token' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_my_profile","arguments":{}}}'

License

MIT