Connect Claude to LinkedIn: MCP Setup and First Prompts
Ruz, Founder LeadShark 路 Jul 7, 2026 路 Updated Nov 11, 2026 路 14 min read
To connect Claude to LinkedIn, add LeadShark's Apex MCP as a custom connector in Claude.ai. Paste https://apex.leadshark.io/mcp, sign in with LeadShark and approve access. Your LeadShark account needs Pro+ or Apex. There is no local LeadShark package or API key to install for this hosted connection.
Once connected, Claude can read your posts and engagers, schedule posts and manage comment-to-DM automations. Apex adds commenting and DMs as you, inside the limits you set. ChatGPT and the other assistants below connect to the same hosted server.
A LinkedIn MCP server gives an AI agent a set of LinkedIn tools it can call through the Model Context Protocol. You talk to your assistant in plain language, and it picks the right tool. LinkedIn doesn't offer an official MCP server for acting from your personal profile, so every LinkedIn MCP server is built by a third party, and Apex is LeadShark's.
Looking for the product page with the server URL and what Apex MCP can do? See LinkedIn MCP Server. This guide is the how-to.
That part is easy to build. The hard part is everything around it: whose limits apply, how fast the agent is allowed to move, what needs your approval, and how you stop it. This guide starts with the setup, then covers all of that.
How to connect Claude to LinkedIn
Claude.ai connects over OAuth. There's nothing to install and no API key to copy.
Create a LeadShark account and connect your LinkedIn account in LeadShark first. Your assistant connects to LeadShark, not to your LinkedIn login. The hosted MCP needs Pro+ or Apex; check current plan access if the connector authorizes but tool calls are refused.
- In Claude.ai, open Customize, then Connectors, click the + next to Connectors and choose Add custom connector.
- Name it (for example "LeadShark Apex") and paste the server URL:
https://apex.leadshark.io/mcp. Leave the optional OAuth fields blank and click Add. - Click Connect. A LeadShark sign-in opens; log in, review the requested scopes (
mcp:readandmcp:write) and click Approve. - Open the connector's Configure view. Set the read-only tools to Always allow and leave the write tools on Ask every time.
- Test it: "Show me my recent LinkedIn posts."
Your first Claude LinkedIn prompts
Start with a read before asking Claude to act:
Show my recent LinkedIn posts and who engaged with the latest one.
Then ask for a draft you can review:
Draft a resource post for my audience. Show me the text and keyword invitation before scheduling anything.
Keep write tools on Ask every time. On Apex, you can also use approval mode to hold writes for review. Connecting the server is not a reason to give an assistant open-ended permission to message people.
Can I connect LinkedIn Pages to Claude?
The MCP docs include list_companies for the Company Pages you manage, and scheduling as a Company Page on Pro+ and Apex. Company Page commenting and reactions are Apex features. Check the account and Page before approving a write; managing a Page and posting as your personal profile are different actions.
Why is my Claude LinkedIn connector missing tools?
Check the LeadShark plan first: Pro+ is the baseline, and Apex adds more tools. If a tool was added or changed, refresh the connector's tool list in Claude's Connectors settings. The MCP docs are the current reference for tool names and plan access.
Quick setup: Claude, ChatGPT, Cursor and Grok Bot
| Assistant | How it connects | What you paste |
|---|---|---|
| Claude.ai | Custom connector, OAuth sign-in | https://apex.leadshark.io/mcp |
| ChatGPT | Custom MCP connector, OAuth sign-in | https://apex.leadshark.io/mcp |
| Grok | Custom connector with sign in | https://apex.leadshark.io/mcp |
| Grok Bot | Custom server added in chat | The Grok Bot request below |
| Cursor | Hosted server with sign in | The URL configuration below |
| Claude Code | Hosted server with sign in | The Claude Code command below |
| Codex | Hosted server with sign in | The Codex command below |
The account needs Pro+ ($59/mo, Automation + Discover MCP with 34 tools) or Apex ($119/mo, the full Apex MCP with all 55 tools). Full steps for each are below. The complete reference is in the Apex MCP docs.
Fastest way
Paste this into an agent that can read a URL:
set up LeadShark: https://apex.leadshark.io/mcp/setup.txt
Follow the steps for your assistant and approve access on apex.leadshark.io.
Grok
- Open https://grok.com/connectors.
- Click New Connector, then choose Custom.
- Paste
https://apex.leadshark.io/mcpas the server URL. - Sign in on apex.leadshark.io and approve access.
Grok Bot
- Ask: "Add a custom MCP server called LeadShark at
https://apex.leadshark.io/mcp". - Confirm with "Add it".
- Click Authorize on the connect card.
- Sign in on apex.leadshark.io and approve access.
Use the words "custom MCP server" so Grok Bot adds your server instead of looking for a catalog plugin.
What can Claude do on LinkedIn?
On its own, Claude can't touch LinkedIn. It has no access to your account until you connect a LinkedIn MCP server. Once you paste https://apex.leadshark.io/mcp into Claude and approve, Claude can work on your LinkedIn account:
- Read: your recent posts and their stats, and everyone who commented, reacted or reposted (Pro+). A person's recent posts, comments and reactions, your home feed and your inbox (Apex).
- Research: search people and companies, find fresh lead-magnet posts in your niche, enrich a profile or a company and judge who fits your ideal customer (Pro+). Search posts by keyword across LinkedIn (Apex).
- Act as you: send connection requests and accept or ignore invitations (Pro+). Comment, react, reply to comments and send DMs (Apex).
- Run your lead magnets: create and edit comment-to-DM automations with follow-up DMs (2 on Pro+, 3 on Apex), schedule or shift posts, build lead pages and pull the emails they captured (Pro+).
Every write draws from the same daily limits your automations use. On Apex, Claude can check what's left, lower a cap or pause a channel, and raising a cap waits for your approval. In Claude, set the write tools to "Ask every time" and nothing it sends goes out without your yes.
ChatGPT gets the same tools through the same URL. The setup steps for both are below.
What is a LinkedIn MCP server?
MCP (Model Context Protocol) is an open standard for connecting AI assistants to outside tools. An MCP server publishes a list of tools, each with a name, a description and the inputs it accepts. The assistant reads that list and decides which tool to call to get your request done.
A LinkedIn MCP server is one where those tools act on LinkedIn. So instead of opening five tabs to find prospects, read their recent posts, check whether you've talked before and write a note, you say:
Find heads of RevOps at B2B SaaS companies in my 2nd-degree network, show me what they've commented on lately, and draft a connection note for the three best fits.
The agent brings the reasoning. The MCP server brings the hands.
Is there an official LinkedIn MCP server?
As of October 2026, LinkedIn doesn't offer an official MCP server for acting from your personal profile. Every "LinkedIn MCP" you'll find is third-party, and they differ mostly in one thing: what they do with your account and how much control you keep over it.
Why a "governed" LinkedIn MCP matters
A connector gives your assistant tools that can affect your account. Before enabling writes, check whose limits apply, whether you can require approval and how you can pause the agent. The question is not just "what can it do?" but "what stops it?"
Here's what governance looks like in LeadShark's Apex MCP:
- Your limits, shared with your automations. DMs, comment replies and daily/weekly connection requests all draw from the same caps you set in the dashboard. An agent can't route around them.
- The agent can't raise its own budget. Through
manage_activity_limitsan agent can check what's left, lower a cap or pause a channel immediately. Raising a cap is staged for the account owner to approve. - Pacing. Every write is velocity-paced and honors a shared cooldown, so actions are spread out instead of fired in a burst.
- Off-hours. If you've turned on off-hours, the agent won't send DMs, comments, reactions or connection requests between 10 PM and 6 AM in your timezone. Reads still work.
- Approval mode (Apex). Put the account in queue mode and every write is staged for your review instead of sent. Nothing reaches LinkedIn until you release it.
- Emergency Stop. One switch (in the dashboard under Agent, then Logs) blocks every agent write on the account instantly: automations, MCP, everything. Reads keep working until you resume.
- Do-not-engage is enforced. Archive a lead and DMs to them are refused at the send step, not just "discouraged" in a prompt.
- A full audit log. Every governed action is logged. Your agent can read its own history with
list_actions(what it did, what was refused and why, what's pending approval), and you can see the same trail in the dashboard.
There's also a client-side layer. In Claude you can set read-only tools to "Always allow" and leave the write tools on "Ask every time", so Claude checks with you before each comment, connection or message.
None of this makes LinkedIn automation risk-free. It means the agent operates inside rules you set, with pacing and an action record. For background, see cloud-based LinkedIn automation.
What an AI agent can do with a LinkedIn MCP server
The Apex MCP has 55 tools. They follow a simple loop: see (find people and posts), decide (enrich, check history, judge fit) and act (comment, connect, message), plus the automation and scheduling tools LeadShark was built on.
Plan access matters here, so I've marked it per row. Pro+ is the MCP baseline; most tools that read other people's activity or act as you are Apex.
| Capability | Example tool(s) | Plan |
|---|---|---|
| Search people or companies | search |
Pro+ |
| Search posts by keyword, by anyone | search_posts |
Apex |
| Find fresh lead-magnet posts in your niche | discover_lead_magnets |
Pro+ |
| Read a person's recent posts, comments and reactions | list_person_posts, get_person_activity |
Apex |
| Turn a post's commenters and reactors into a lead list | list_post_engagers |
Pro+ |
| See your hottest inbound leads | list_signals |
Apex |
| Enrich a person or company profile | enrich, enrich_company |
Pro+ (email lookup is Apex) |
| Pre-DM briefing: your shared history with someone | get_lead_activity |
Apex |
| Comment on or react to a post as yourself | comment_on_post, react_to_post |
Apex |
| Reply to or like a comment | engage_with_comment |
Apex |
| Send a connection request | send_connection_request |
Pro+ |
| Accept or ignore inbound invitations in bulk | manage_invitations |
Pro+ |
| Send a DM (or InMail, using your own credits) | send_message |
Apex |
| Read your inbox and a conversation | list_recent_messages, get_messages_with_person |
Apex |
| Schedule, edit, shift or cancel posts | schedule_post_with_automation, edit_scheduled_post, shift_scheduled_posts, cancel_scheduled_post |
Pro+ |
| Create and manage comment-to-DM automations | create_automation, edit_automation, list_automations |
Pro+ |
| Build lead pages and pull captured emails | manage_lead_pages, get_page_leads |
Pro+ |
| Pause DMs instantly | set_daily_dm_limit |
Pro+ |
| Check remaining limits, pause or tighten any channel | manage_activity_limits |
Apex |
| Read the agent's own action log | list_actions |
Apex |
Enrichment runs through your own connected LinkedIn account and is limited to 250 lookups a day, so save it for leads worth pursuing. With send_message, a normal DM goes to a connection; reaching a non-connection uses InMail and spends your LinkedIn plan's paid credits. Check the recipient and the proposed action before approving.
The full, current tool reference with inputs and outputs is in the MCP docs.
Example prompts for your agent
These are the kinds of requests that work well. You don't need tool names; the assistant figures that out.
Who engaged with my last post? Build me a lead list, warmest first.
Find recent LinkedIn posts about hiring SDRs and like the three most relevant.
What has Jane Doe been commenting on lately? Draft a connection note
that references it, and check we haven't talked before.
Show my pending connection requests. Accept the founders, ignore the rest.
Schedule my guide post for Tuesday 9am and DM anyone who comments "guide".
I'm off next week. Push all my scheduled posts back 7 days.
What have you already done today? Anything refused or waiting for approval?
Set my daily DM limit to zero.
The comment-to-DM flow in the middle is the same lead-magnet automation the dashboard runs. If that's new to you, here's how turning post comments into automated DMs works.
How to connect it to ChatGPT
In ChatGPT, add LeadShark as a custom MCP connector, paste your server URL (https://apex.leadshark.io/mcp) and sign in with OAuth. Review the scopes and click Approve. The menu differs by plan, and so does which plans can use write tools from a custom connector.
How to connect it to Cursor, Claude Code or Codex
All three connect to the hosted server at https://apex.leadshark.io/mcp. No local LeadShark package or API key is needed. Prefer REST calls from your own code? See the LinkedIn API guide.
Cursor
Add this to .cursor/mcp.json for your project or ~/.cursor/mcp.json for all projects:
{
"mcpServers": {
"leadshark": {
"url": "https://apex.leadshark.io/mcp"
}
}
}
Open Cursor Settings, then MCP, click Connect next to leadshark and sign in on apex.leadshark.io. Do not add an API key to this file.
Claude Code
Run:
claude mcp add --transport http leadshark https://apex.leadshark.io/mcp
Then run /mcp, select leadshark and sign in on apex.leadshark.io.
Codex
Run:
codex mcp add leadshark --url https://apex.leadshark.io/mcp
Sign in when the browser opens. If it does not open, run codex mcp login leadshark. Check the connection with codex mcp list.
Whichever client you use, the account needs Pro+ or Apex. On any other plan the server refuses tool calls.
Which plan do you need?
- Pro ($39/month) covers LeadShark's lead-magnet automations in the dashboard, but not the MCP.
- Pro+ ($59/month) connects the MCP with the baseline tools: automations, scheduling, lead pages, people and company search, post engagers, connection requests, invitation triage, lead-magnet discovery and enrichment.
- Apex ($119/month) adds the full agent loadout: post search, activity reads, inbox, signals, and acting as yourself (comments, reactions, comment replies, DMs), plus the limits dashboard and action log.
New accounts get a 7-day free trial with no card. On top of that, any account not already on Apex can unlock a one-time, 72-hour Apex trial from Settings to try the full MCP with their own assistant. Compare plans on the pricing page.
Try it with your own agent
If you already live in Claude, ChatGPT or Cursor, the quickest way to judge whether a LinkedIn MCP server is worth it is to connect one and ask it something real. Start the 7-day free trial, no card: create a LeadShark account, then unlock the one-time 72-hour Apex window in Settings and paste the server URL into your assistant. Ask it who engaged with your last post. Keep the write tools on "ask every time" for the first week and watch what it proposes.
FAQ
Can Claude post on LinkedIn for me?
Yes, once you connect a LinkedIn MCP server. With LeadShark, Claude can schedule your posts with a comment-to-DM automation attached on Pro+, and comment, react and DM as you on Apex, all inside the daily limits you set.
Does ChatGPT work with LinkedIn?
Not on its own. ChatGPT needs a LinkedIn MCP server added as a custom MCP server. LeadShark's uses the same URL as Claude, https://apex.leadshark.io/mcp, with a LeadShark sign-in, and works on Pro+ and Apex.
Is it safe to connect Claude to my LinkedIn account?
It's safer when the server is governed: shared daily limits, human pacing, off-hours, an action log and an Emergency Stop, plus an approval mode on Apex that holds every write for your review. Nothing makes LinkedIn automation risk-free, so keep the write tools on "Ask every time" while you get used to it.
Does LinkedIn have an official MCP server?
As of October 2026, LinkedIn doesn't offer an official MCP server for acting from your personal profile. Every LinkedIn MCP server available is built by a third party. When comparing them, look at how they treat your account: limits, pacing, approvals and logging.
How do I connect Claude to LinkedIn?
Add a LinkedIn MCP server as a custom connector. In Claude.ai, open Customize, then Connectors, choose Add custom connector, paste https://apex.leadshark.io/mcp, then sign in with LeadShark and approve. The LeadShark account needs Pro+ or Apex, and there's nothing to install.
How do I connect ChatGPT to LinkedIn?
In ChatGPT, add LeadShark as a custom MCP connector, paste your server URL (https://apex.leadshark.io/mcp) and sign in with OAuth, then approve with your LeadShark account. The menu differs by plan. The account needs Pro+ or Apex.
What should I look for in a LinkedIn MCP server?
Look at how it treats your account, not just how many tools it has. Check whose limits apply, whether actions are paced, whether writes can wait for approval, and whether there is an action log and Emergency Stop. Apex MCP provides these controls, uses your own account with your consent and keeps sends inside your daily limit of 0 to 250 DMs a day (150 recommended max).
Can I use a LinkedIn MCP server with Claude?
Yes. In Claude.ai, add a custom connector under Customize, then Connectors, paste https://apex.leadshark.io/mcp and sign in on apex.leadshark.io to approve access. For Claude Code, run claude mcp add --transport http leadshark https://apex.leadshark.io/mcp, then run /mcp, select leadshark and sign in. The account needs Pro+ or Apex.
Is using an AI agent on LinkedIn safe for my account?
It depends on the controls between the agent and your account. Apex MCP shares your account limits, paces actions, respects off-hours, offers approval mode on Apex and has an Emergency Stop. No automation is risk-free. Review the proposed action and keep writes on Ask every time while you learn the workflow.
Can the agent send messages without my approval?
You control that through permissions and approval settings. On Apex, approval mode can hold every write for review. In Claude, Ask every time requires confirmation for each write tool. Sends still count against the same account limits as your automations, and DMs to archived leads are refused.
What's the difference between Pro+ and Apex for MCP?
Pro+ connects the MCP and covers your own content and pipeline: automations, scheduling, lead pages, search, post engagers, connection requests, invitation triage, discovery and enrichment. Apex adds the tools that read other people's activity and act as you: post search, activity and inbox reads, signals, and commenting, reacting and DMing as yourself.
Do I need an API key?
Not for the hosted setup described here. Claude.ai, ChatGPT, Grok, Grok Bot, Cursor, Claude Code and Codex connect to https://apex.leadshark.io/mcp and sign in on apex.leadshark.io. No local LeadShark package or API key is needed.

Written by
Ruz
Founder LeadShark
Put your lead magnets on autopilot
Reply to every commenter and deliver your resource in DMs, automatically. 7-day free trial, no card.
Start automating for free