Skip to content

LinkedIn MCP servers: what exists and what is allowed

5 min read

PostCat AI is not affiliated with LinkedIn.

A LinkedIn MCP server is a program that lets an AI assistant work with LinkedIn through tools, such as looking up a contact. LinkedIn's API terms restrict access to its content outside the official APIs, and they bar scraping and automated posting. Before you connect one, check what it asks for.

Artem Luko

Founder of PostCat AI

0+

followers on LinkedIn

He grew his own LinkedIn post by post. PostCat AI is everything he learned, in one place.

Learn how

What an MCP server is

MCP stands for Model Context Protocol. It is a way for an AI assistant to call outside tools. An MCP server is the part that offers those tools.

Microsoft's documentation describes one example this way: "The server exposes various tools that let AI agents look up contacts by name, email, or LinkedIn URL." That example's free tools use no Universal Credits. It keeps your credit balance and the costs of paid actions visible, and it asks for your consent before a paid action. The details are on Microsoft Learn's connector page.

Read the tool list before you connect anything, so you know what the tool can reach.

Where you meet one

The place to look is the consent screen. LinkedIn's Help Center says that before you connect your LinkedIn account to a third-party application, you'll see details of what LinkedIn data you'll be sharing. That data may be used to create your account on the application or to make your experience on the application more personalized. You may also authorize the application to take actions on your behalf, such as sharing content and sending messages.

Read that list before you approve. Consent does not last forever. Upon your consent, a third-party application will have access to your LinkedIn data for up to 1 year before you will need to re-consent. If you want a copy of your data without a live connection, you can download an archive of your LinkedIn data in your LinkedIn settings.

Be careful with any tool that asks you to type your LinkedIn username and password into it directly. LinkedIn's Member Portability terms bar functionality that collects LinkedIn usernames or passwords, so treat that request as a warning sign.

What LinkedIn's API terms do not allow

These rules are written for people who build and run apps. A careful tool should keep to the same lines. Under LinkedIn's API Terms of Use, the list includes:

  • Accessing, storing, displaying or transferring LinkedIn content obtained by automated data collection, crawling, spidering or other technology outside the APIs, including content a third party got for you.
  • Using the Content or the APIs to automate posting on the LinkedIn Services.
  • Accessing a Member's network of connections through your Application without their express permission.
  • Trying to exceed or circumvent limitations on API calls and use, including by creating multiple Applications for identical, or largely similar, usage.
  • Requiring your Users to obtain their own Access Credentials to use your Application.
  • Distributing or allowing third parties access to any stand-alone API.
  • Making statements that express or imply your Application is sponsored or endorsed by LinkedIn.
  • Obfuscating or hiding LinkedIn buttons, sign-in functionality, or consent or authorization flows from your Users.
  • Using any API in a manner that LinkedIn, in its reasonable discretion, considers abusive.

The Member Portability API terms also bar functionality that proxies, requests or collects LinkedIn usernames or passwords.

How a developer gets API access

If you build your own tool, the terms start with sign-in and registration:

  1. Sign in to LinkedIn with your personal or corporate LinkedIn account username and password.
  2. Click My Apps and register your Application by following the instructions.

After that, the registration details must stay accurate. Your Application's name and description, and your Member account, including your current title, company and e-mail address, must be accurate and up to date at all times.

The APIs have limits you need to plan for. Your Application is not expected to have more than 100,000 lifetime Users. It must not attempt to exceed any applicable API call limits, which LinkedIn sets in its sole discretion, or 100,000 daily calls where no use-quota restriction applies. Your use of the APIs must comply with the technical documentation, usage guidelines and call volume limits maintained at the Developer Site.

What LinkedIn's documentation does not say

LinkedIn's documentation does not state these points, so do not assume them:

  • Whether LinkedIn offers its own MCP server or connector.
  • Which LinkedIn plans (free, Premium, Sales Navigator) can use third-party connections.
  • The exact settings path for managing connected apps, on desktop or mobile.
  • What LinkedIn says to do when an account is flagged or restricted after automated activity.
  • The steps, plan or country limits for adding a LinkedIn connector in an AI assistant.
Sources
  • API Terms of Use (linkedin.com)
  • LinkedIn Help: Third-party applications data use
  • LinkedIn DMA Portability API Terms (linkedin.com)
  • LeadIQ MCP - Connectors (learn.microsoft.com)
  • LinkedIn Research API Program Terms (linkedin.com)
  • LinkedIn Help: Third Party

Frequently asked questions

LinkedIn's documentation does not say whether it offers its own MCP server or connector. Check who publishes the tool and what it says about its access to LinkedIn. Be wary of any tool that implies LinkedIn sponsors or endorses it, because the API terms bar that kind of statement.

Start with your team's next post.

Connect your team's LinkedIn profiles and start with a free trial.