ProspectingConnector
Search LinkedIn's people and company graph, send connection requests and messages as yourself, from Claude or ChatGPT through Tulina's own MCP server.
Official site: www.linkedin.com
What it can do
2 read/2 send
- 01Read
Search people and companies across LinkedIn's graph, on Tulina's own access — nothing to configure
- 02Read
Search, read profiles and browse your network under your own LinkedIn session, including Sales Navigator or Recruiter when your account has it
- 03Send
Send connection invitations and messages, and reply in the thread
- 04Send
Read your LinkedIn feed and comment on a post
Two different things sit behind one name here. Search runs on Tulina's own platform access to AI Ark — shared capacity, nothing for a workspace to key in. Messaging runs through Unipile's hosted per-person session: each person connects their own account, and it acts as them, not as a shared bot.
That split is not an implementation detail to hide. Sending an invitation or replying in a thread happens under someone's real identity — worth knowing which of the two you are pointing a model at.
If you are choosing between this and a dedicated sequencer, LinkedIn automation tools compared sorts the usual ones by where they run and what LinkedIn's rules say about all of them. To write the messages with Claude or ChatGPT and keep every reply answered, see LinkedIn outreach with Claude or ChatGPT.
Does LinkedIn have an official MCP?
We found no MCP server published by LinkedIn itself as of October 2026. The LinkedIn MCP servers that come up for the query are third-party: open-source projects that drive a browser with your session cookie, data providers that sell LinkedIn-sourced records, and automation platforms. Tulina's LinkedIn connector is one of them, built on two providers it names on this page.
What is the LinkedIn MCP server URL?
We know of no LinkedIn-hosted URL to add. Through Tulina, LinkedIn rides
on the same link as every other connected tool, https://mcp.tulina.ai/mcp,
which you add to Claude or ChatGPT once. The MCP page has it.
How do you connect LinkedIn to Claude and ChatGPT?
Through Tulina's own MCP server. Search people and companies on Tulina's own access, then send a connection request or message under your own LinkedIn session, and reply in the thread — from the same chat as Claude or ChatGPT.
- Add the Tulina link to Claude or ChatGPT and sign in. Search through AI Ark works from here, with nothing else to set up.
- To act as yourself, click "Connect my LinkedIn account" on Tulina's LinkedIn connector. There is no cookie to paste and no browser extension; the session runs on a hosted browser that Unipile operates.
- If you use Sales Navigator or Recruiter, attach it when you connect. An account carries one of the two, and adding it later is a reconnection.
Two ways in: AI Ark search or your own session
| AI Ark search | Your LinkedIn session | |
|---|---|---|
| What it reads | A purchased index of LinkedIn-sourced people and companies | What your own account sees: network, profiles, posts, feed, jobs |
| Setup | None; it runs on Tulina's key and draws on workspace credits | Connect your account once |
| Acts on LinkedIn | Never | Sends invitations, messages and comments as you |
| Limits | Billed per returned record | LinkedIn's own rate limits on your account |
A list of two hundred finance directors at mid-sized companies is a job for AI Ark. Reading a prospect's last posts before writing to them, and then writing to them, is a job for your session through Unipile.
What you can ask it
"Find founders of companies of 50 to 200 people, based in Germany." The agent searches AI Ark for people, filtering the company by size and the person by seniority and location, and returns names, titles and companies. It filters the person's location rather than the company's, because AI Ark reads a company's location as its headquarters.
"Which of my first-degree connections work in fintech in Paris?" The agent searches under your own session, limited to first-degree connections, the Paris area and the industry, and returns the profiles. The results are what your account can see, so they differ from a colleague's.
"Invite the three people from Acme on that list, with a short note, and reply when they accept." The agent sends each invitation from your account with the note, then answers in the thread once a person accepts. Sending, inviting and publishing are separate tools from reading, so a process can be allowed to search without being allowed to write.
Follow champions who change jobs, Turn website visitors into outreach and Social listening to qualified leads all run partly on LinkedIn.
What the LinkedIn connector can't do
- Return an email address or a mobile number a profile does not publish. That takes an enrichment connector such as FullEnrich.
- Go past LinkedIn's limits on your account. A burst of calls is met with a wait that grows with your recent pace, up to about an hour, and pushing through it can get the session disconnected. Volume is paced.
- Count a population on an ordinary account. Classic search gives no total and does not page through results; Sales Navigator and Recruiter page further.
- Use Sales Navigator and Recruiter on one account at once.
- Send anything as a shared company bot. Every message and invitation leaves under the person whose session it is.
Why through Tulina
The connection was never the hard part.
Most systems now publish their own way for a model to reach them, and that is good: it is what Tulina connects to. The hard part is deciding who may use it, what it is allowed to return, and what the model knows before it asks.
The keys stay in one place
YouTeamOrganisation
Connect LinkedIn once. The credential is encrypted in a vault, never held in clear, and never pasted into a chat or a workflow file. A person uses their own key where they have one, their team's where they do not, the organisation's after that, so nobody is blocked and nobody holds more access than they need. Revoke it once and everything running on it stops.
What leaves is filtered before it leaves
fields droppedmasked or hashedpseudonymised
Roles decide who can reach LinkedIn at all. Beyond that, a policy on the connector itself decides what comes back. The data is minimised at the source rather than after it has already reached a model.
Every call is on the record
whatwhowhen
Runs are logged for the organisation. That is what makes an automated action reviewable after the fact, and it is what a security review asks for first.
The run starts knowing your company
namingkey accountsdecisions
A direct connection gives a model access. It does not tell it how your company names things, which accounts matter, or what was decided last quarter. Tulina carries that across every tool, so the same request against LinkedIn returns something usable instead of something generic.
Processes that run on LinkedIn
CRM hygiene and prospect job changes
Weekly CRM data cleansing: duplicates merged with their history kept, prospects who changed jobs moved to their new company, and a recap posted to the team.
4 steps →Follow champions who change jobs
Each week, champions at won accounts are checked for a new employer, the new company is scored on your ICP, and a warm deal opens with a note for the right rep.
7 steps →Message new LinkedIn connections
Each new LinkedIn connection is checked for past conversations, researched on public sources and given a first message and a follow-up, sent once approved.
9 steps →Prepare for a demo from the last call
The last calls, the deal's state and fresh research become one prep brief, plus a deck built on that company's own use cases, posted before the demo.
5 steps →Reply to every LinkedIn conversation
Four times a day, LinkedIn threads where the other person wrote last come back to you in Slack with a reply drafted in your style. Nothing is sent for you.
9 steps →Revive deals that went cold
Deals lost more than 90 days ago come back with fresh signals, contacts checked for job moves, and the features they asked for that have since shipped.
6 steps →Run LinkedIn outreach at a safe pace
Small sending slots through the day, a daily budget that climbs one step only while the account stays healthy, and every send stopped at the first warning.
7 steps →Social listening to qualified leads
Every morning, competitor-post engagers and companies reading your site are matched to the CRM, qualified on your ICP, ranked by intent and handed to their rep.
7 steps →Source and qualify leads
A target company gets searched across LinkedIn, ranked against your ICP, and the qualified shortlist lands in your CRM with email and mobile filled in.
6 steps →Turn signups into pipeline
PQL scoring for every trial signup: enriched, scored on your ICP, the first session read, a play chosen by your rules and a deal opened for the right rep.
7 steps →Turn website visitors into outreach
Website visitor tracking that ends in outreach: each company is checked against your CRM, and the few with real intent get a named contact and a drafted note.
7 steps →
