Live. Included in Growth and Pro, or from $29 on its own

The ContentIn LinkedIn MCP Server

The ContentIn LinkedIn MCP connects Claude, Claude Code, ChatGPT, Cursor, VS Code or any MCP client straight to LinkedIn. Publish, schedule and read your analytics from the chat — and write posts in your own voice with the ContentIn ghostwriter you already know. Included in Growth and Pro, or on its own with ContentIn for Agents from $29. One connection per profile.

One click to connect: paste the server URL into your client and sign in with ContentIn. No key to copy.

Compatible with the MCP clients you already use

  • Claude.ai
  • Claude Code
  • Claude Desktop
  • ChatGPT
  • Cursor
  • VS Code
  • Any MCP-compatible client

One remote server, connected once. Compatibility comes from the protocol rather than a plugin per app: every client that speaks Streamable HTTP MCP gets the same set of tools, with nothing to install. Step-by-step setup for Claude, ChatGPT, Cursor and VS Code is further down this page.

Running your own agents or automations for LinkedIn? See LinkedIn for your AI agent.

What is a LinkedIn MCP server?

MCP (Model Context Protocol) is the open standard that lets AI assistants like Claude securely connect to external tools and data. A LinkedIn MCP server exposes LinkedIn as a set of tools your AI can actually use — so instead of describing what you want and copy-pasting the result, your assistant publishes posts, reads engagement data, and manages your content directly. For the plain-language definition, read our LinkedIn MCP glossary entry.

The ContentIn LinkedIn MCP is remote and connects with one click. You add the server URL to your AI client, sign in with ContentIn, pick the profile the connection acts for and approve — scoped to that single profile, revocable from the Connections tab of your ContentIn profile at any moment. Your LinkedIn password never touches it, and the two tools that can put something in front of an audience — schedule_post and publish_post — always show you the exact text and wait for your confirmation before anything happens.

The tool surface

What an MCP server for LinkedIn lets your assistant do

Thirteen tools, one connection. Included in Growth and Pro, or on its own with ContentIn for Agents from $29 (the leads and comments tools need Pro or Agent Pro). Scheduling and publishing also work without a plan, up to 2 posts a day and 30 a month per LinkedIn profile. Some wire LinkedIn itself into your AI — publishing, scheduling, analytics, the people engaging with you. The rest wire in your ContentIn ghostwriter: your voice, your substance, your rules.

Only in ContentIn

Posts that sound like you — not like AI

write_post_in_my_voice

ContentIn learns your voice from the posts you've actually written: your hooks, your rhythm, the phrases you keep coming back to. Brain-dump into your AI client and get back a draft you'd put your own name on.

Only in ContentIn

Save a story to ContentIn without leaving your AI

capture_substance

Hit a story worth telling, a strong opinion, a number you keep quoting? Tell your AI to save it and it lands in your substance bank. Nothing is saved unless you ask. Every future draft can draw on it.

The ghostwriter layer

Everything ContentIn does today, callable from your AI client. It runs on the model we've trained on your writing — which is the part a generic LinkedIn wrapper can't copy.

Let your own AI write like you

get_voice_context

Your voice, your rules, your saved stories, opinions and numbers and your best posts, served to your assistant before it writes. Keep the model you like; it drafts from your real material instead of inventing it. Read-only: no AI writing on our side, nothing goes out to LinkedIn.

Never open a blank chat again

generate_ideas

Ideas drawn from your topics, your audience and what's already resonated — not generic prompts anybody could have gotten.

Turn one good post into another

repurpose_post

Take something that landed and rework it from a fresh angle that still sounds like you.

Publish & schedule

Your LinkedIn account, exposed to your assistant as tools it can actually call — so the draft you just approved doesn't need a copy-paste to go anywhere.

Check your setup in one ask

get_account_status

Your setup, plan and what each tool needs. Works on any connected account, with or without a plan, so your assistant can tell you what is missing instead of guessing.

Publish the moment it's ready

publish_post

Your assistant shows you the final text, you confirm, it goes live. Nothing is ever published without your explicit yes.

Queue it for the right moment

schedule_post

Put a post on the calendar for a specific time, without leaving the conversation — same confirmation step, nothing queued behind your back.

Send the picture with the post

attach_images

Attach up to 5 images to a draft — a public image URL or the file itself — then publish or schedule it as usual. JPG, PNG or WEBP, 8 MB each; one image goes out as a single image, two or more as a LinkedIn multi-image post. No GIFs, video or PDFs. Fair use: 30 images an hour, 100 a day per key.

See your whole pipeline in one ask

list_posts

Drafts, scheduled and published posts, straight into the chat. No tab-switching to check what's queued.

Analytics & audience

The feedback loop, in the same conversation that writes the next post, so your AI can reason about performance, and about the people behind it, instead of guessing at either. Post analytics are included in Growth and Agent; the leads and comments tools need Pro or Agent Pro.

Ask why a post worked

get_post_analytics

Impressions, members reached, reactions, comments and shares per post — the raw evidence behind what landed and what didn't.

See who's actually engaging

list_leads

Pro or Agent Pro. Everyone who reacted to, commented on or shared your posts: name, headline, profile, ContentIn's ICP fit score, lead status and the posts that pulled them in. Filter by status, score or last interaction. A single reaction is enough to create a lead; an unscored one means not scored yet, not a poor fit.

Find the comments you left

list_my_comments

Pro or Agent Pro. The comments you wrote on other people's posts, with a link back to each one, what it got, and the post it sits under. Read from a daily sync, so the last day can still be missing, and only the counts, never who reacted.

Every tool here is included in Growth and Pro, or on its own with ContentIn for Agents plans from $29. list_leads and list_my_comments need Pro or Agent Pro. schedule_post, publish_post and attach_images also work on an account with no plan, up to 2 posts a day and 30 a month per LinkedIn profile. One connection per profile. Your assistant picks the right tool from what you ask, so you never call them by name. Connecting takes two minutes; the steps are just below.

Included in Growth and Pro

Your ghostwriter, wired into your AI tools — for every profile you run

ContentIn's LinkedIn MCP server is included in the Growth and Pro plans, so web app customers have nothing to buy; the leads and comments tools are Pro. Don't need the web app? Get the same tools on their own with ContentIn for Agents from $29. It's scoped per profile: you pick the profile when you connect, so the assistant you're talking to is always working with that profile's voice, substance bank and calendar — and if you run clients, each client profile gets its own connection, which makes it natural to keep every client in a dedicated AI project.

  • Included in Growth and Pro

    Not an add-on for web app customers: the MCP server is included in Growth, and the leads and comments tools come with Pro. Without the web app, ContentIn for Agents lets your AI schedule and publish for free, and its plans start at $29.

  • One connection per profile

    You pick the profile when you connect: its voice model, its substance bank, its queue. Revoke it any time.

  • Built for running several

    Agency or multi-brand? Connect each client profile where you actually work on it.

Two-minute setup

Get started with the ContentIn LinkedIn MCP in two minutes

The same remote server works in claude.ai, Claude Desktop, Claude Code, ChatGPT, Cursor and VS Code. Nothing to install and no key to copy — add the URL, sign in with ContentIn, and your assistant is working with your profile.

  1. 1

    Add the server URL to your client

    The LinkedIn MCP server URL is the same for every client. No header, no key — just the URL.

    Server URL
    https://mcp.contentin.io/mcp-server
    Sign-in
    ContentIn (OAuth), no key needed

    In claude.ai and Claude Desktop, open Settings → Connectors → Add custom connector and paste the URL. In Claude Code it's one terminal command: claude mcp add --transport http contentin https://mcp.contentin.io/mcp-server. In ChatGPT, Cursor and VS Code, add it as a new MCP server or connector with just the URL.

  2. 2

    Sign in with ContentIn and pick a profile

    Your client opens a ContentIn sign-in page. Log in, choose which profile the connection acts for, and approve. That's it — the connection is scoped to that one profile, and you can revoke it any time in the Connections tab of your ContentIn profile.

    Running a CLI or an agent that can't open a browser? Create an API key instead and send it in the X-MCP-Key header — the API key steps are in the setup guide.

  3. 3

    Try a first prompt

    Once you've approved, ask your assistant something real. Your assistant chooses the right tool from the thirteen above, so you never call them by name.

    “What have I got scheduled for next week, and what did my last three published posts do?”

Per-client walkthroughs, the full tool list with example prompts, and troubleshooting live in the LinkedIn MCP setup guide.

Frequently asked questions

What is the LinkedIn MCP server URL?
ContentIn's remote LinkedIn MCP server URL is https://mcp.contentin.io/mcp-server. Add that URL in claude.ai, Claude Desktop, Claude Code, ChatGPT, Cursor, VS Code or any Streamable HTTP MCP client — no header and no key in the URL. Your client opens a ContentIn sign-in page, you pick the profile and approve. CLIs and agents that can't open a browser can use an API key in the X-MCP-Key header instead. There is no LinkedIn-hosted MCP URL — every LinkedIn MCP server is run by a third party, and this one is ours.
Is the ContentIn LinkedIn MCP compatible with Claude?
Yes — with claude.ai, Claude Desktop and Claude Code. In claude.ai and Claude Desktop, add https://mcp.contentin.io/mcp-server as a custom connector; in Claude Code, run claude mcp add --transport http contentin https://mcp.contentin.io/mcp-server and authenticate from /mcp. Either way you sign in with ContentIn and pick a profile — no key to copy. The steps are in the Get started section above.
Is the ContentIn LinkedIn MCP compatible with Cursor?
Yes. Cursor speaks the Model Context Protocol over Streamable HTTP, so it connects to the same remote server as every other client: one mcp.json entry with just the server URL, sign in with ContentIn when Cursor asks, and the thirteen LinkedIn tools are in the chat. VS Code works the same way. The steps are in the setup guide linked above.
Is the ContentIn LinkedIn MCP compatible with ChatGPT?
Yes. ChatGPT connects to remote MCP servers in developer mode, and ours is remote, so you add the same server URL as every other client and sign in with ContentIn — there is nothing ChatGPT-specific to install and no key to paste. The steps are in the setup guide linked above.
Is the LinkedIn MCP compatible with any other AI agent?
If it speaks the Model Context Protocol over Streamable HTTP, yes. The server is remote and client-agnostic — no ContentIn plugin or SDK per app — so an MCP-compatible agent connects with the same URL as Claude or Cursor and signs in with ContentIn. Agents and CLIs that can't open a browser use an API key in the X-MCP-Key header instead. We only name the clients we use ourselves; anything else that implements MCP connects the same way.
How do I get started with the LinkedIn MCP server?
Three steps, about two minutes, with a ContentIn account and a connected LinkedIn profile. Scheduling and publishing work without a plan, up to 2 posts a day and 30 a month; the other tools need Growth, Pro, Agent or Agent Pro. First, add https://mcp.contentin.io/mcp-server to your client. Second, sign in with ContentIn when the client asks, pick the profile and approve. Third, ask it something real: list your scheduled posts, write a draft in your voice, or pull last week's analytics.
Can the ContentIn LinkedIn MCP read my LinkedIn analytics?
Yes, through LinkedIn's official API for the LinkedIn account connected to your ContentIn profile. get_post_analytics returns impressions, members reached, reactions, comments, shares and engagement rate per published post; list_leads shows everyone who reacted, commented or shared, with ContentIn's ICP fit score; list_my_comments returns the comments you left on other people's posts and what they earned. The tools cover your personal profile — company-page analytics are not part of the tool surface today.
Does LinkedIn have an official MCP server?
No — LinkedIn doesn't publish an MCP server of its own, so every LinkedIn MCP server you'll find is third-party, including ours, which runs on the official LinkedIn API through your connected account rather than on scraping. For how the available servers differ, see LinkedIn MCP servers compared.
Is it secure?
Yes. You authorise each AI client by signing in with ContentIn and approving it for one profile — the connection never sees your LinkedIn password or your other profiles, and you can revoke it from the Connections tab on the Profiles page at any time. Publishing is a separate permission your client asks for the first time it publishes. Scheduling and publishing are confirmation-gated on top of that: your assistant has to show you the exact text and get your explicit go-ahead before anything reaches LinkedIn.
Which plan includes the MCP server?
Growth and Pro include it, so web app customers have nothing to buy. Publishing, scheduling, post analytics and the ghostwriter tools all work on Growth; list_leads (the people engaging with you) and list_my_comments (the comments you left) need Pro, and on Growth they say so instead of returning data. Don't need the web app? The same tools are sold on their own as ContentIn for Agents: Agent at $29 a month or $276 a year, and Agent Pro (with the leads tools) at $49 a month or $468 a year, per LinkedIn profile, with a 7-day trial. Without a plan, scheduling, publishing and attaching images work on any account, up to 2 posts a day and 30 a month per LinkedIn profile. See ContentIn for Agents pricing. It's scoped per profile: every connection acts for the one profile you pick when you connect, including each client profile if you run an agency.
How is this different from a generic LinkedIn MCP server?
Anyone can wrap the LinkedIn API, and moving text from a chat window to your feed is the easy part. The difference is whose voice writes the post: ours is the only one where the model drafting it has learned from the posts you've actually published — your hooks, your rhythm, the phrases you keep coming back to. And it runs both ways: anything you capture in a conversation — a story, a strong opinion, a rule you never want broken — lands in your substance bank and shapes every future draft, in your AI client and in ContentIn.
Can I try ContentIn's AI writing before I subscribe?
Yes. ContentIn's writing tools need a plan (Growth, Pro, or ContentIn for Agents from $29), but the writing engine behind it is open to everyone. Our free AI LinkedIn post generator drafts posts from your own LinkedIn profile in seconds, no login required, so you can see the voice-first approach for yourself today. The MCP server is that same engine wired into your AI client.

Connect your AI to LinkedIn

The ContentIn LinkedIn MCP is included in Growth and Pro, or on its own with ContentIn for Agents from $29. Start a trial, connect your AI client with one click, and your assistant is writing and publishing in your voice a couple of minutes later.