Zevari
Remote MCP server for LinkedIn sales workflows: research prospects, draft outreach, classify LinkedIn inbox replies, prepare campaigns, and stage sensitive LinkedIn actions behind review gates.
Documentation
The all-in-one LinkedIn MCP server.
A LinkedIn MCP exposes LinkedIn as tools an agent can call: search, score, draft, send, comment, classify. Zevari is the hosted LinkedIn MCP server, and behind the endpoint is a full outbound department: six specialists that take a prospect to a qualified conversation. Run them from Claude, ChatGPT, Codex, Cursor (including Grok Bot in Cursor), Gemini, Perplexity, VS Code or any MCP client, or let them run from the Zevari app. Either way every send waits for your approval until you switch a trained specialist to Autopilot. No browser cookies.
LinkedIn MCPREST APIClaude Code, Codex, CursorApproval gatesNo cookies
claude mcp add --transport http zevari https://mcp.zevari.ai/mcp
Run that, then /mcp inside Claude Code to sign in. Using Claude.ai, ChatGPT or Codex? See every client.
Your agent discovers Zevari's LinkedIn tools through MCP.
Zevari loads workspace context, Voice DNA, and safety rules.
Outbound is drafted and staged for your approval until you switch a trained specialist to Autopilot.
What is a LinkedIn MCP
MCP gives an agent hands. A LinkedIn MCP gives it LinkedIn.
MCP (Model Context Protocol) is the open standard that lets an agent call tools. A LinkedIn MCP exposes LinkedIn as tools: search profiles, score a prospect, draft a connection note, send a message, comment on a post, classify an inbox reply. Without one, an agent can find the needle in the haystack - it just cannot reach in and pull it out. Zevari is the hosted reach: free to add the endpoint, Operator for the platform, and every write approved by you until you switch a trained specialist to Autopilot.
With Zevari, that means a full outbound department for LinkedIn and email your AI can hand work to: Prospecting, Signals, ICP Scoring, Voice Writer, Follow-up and Reply Qualifier. You describe who you want to reach; they find, score, write and follow up. It is a hosted connector with no browser extension and no cookie injection, and a human approves every send until a trained specialist moves to Autopilot.
How Zevari connects an agent to LinkedIn
Hosted, nothing to babysit
Zevari is hosted. There is nothing to self-host, no browser automation to babysit, no cookie to refresh. Add one MCP endpoint, authorize through OAuth, and your agent has 191 LinkedIn and GTM tools. Engineers who prefer to drive it from their own stack call the 109 public REST endpoints.
If the first step in your stack is contact-data enrichment, read the MoltSets for LinkedIn guide and the developer workflow for turning verified emails, mobiles, companies, and LinkedIn URLs into safe Zevari campaigns.
One thing raw Claude Code cannot do alone: persistent scheduling. Claude runs when you run it, then forgets. Zevari holds state on our infrastructure - campaigns advance, warm-up sequences fire, the inbox gets classified - while your agent sleeps. Your agent sleeps; your pipeline does not.
A real session
You: Find 15 founders who posted about hiring an SDR in the last 30 days.
Agent: Searches signals, scores each as Strong, Moderate or Weak fit with reasons, returns 15.
You: Build a 3-step warm-up campaign for the Strong fits. Voice-match it.
Agent: Drafts notes in your Voice DNA, stages a warm-up plus connect sequence.
You: Approve.
Agent: Executes within your weekly ceiling, on a human-paced schedule.
You approve from the chat or from the Zevari app. The specialists do the work you would have to do, and you stay in control of every send.
What your agents can do
Real tools your agent calls, not a roadmap
The tool surface is 191 functions. You do not call them by hand - you tell your agent what you want and it orchestrates. No prompting expertise needed; just talk to it.
Voice DNA
Drafts trained on your sent messages, so outreach sounds like you, not a template and not generic AI. Another version of you that does not read like a robot.
Signal-based targeting
Finds people who posted about a topic in the last 30 days, sorts each one into Strong, Moderate or Weak fit with written reasons, and reaches intent instead of a static list - the needle, not the haystack.
Campaigns and sequences
Multi-step LinkedIn campaigns your agent builds and advances on a schedule. Set the sequence once; targets move through it while you work on something else.
Inbox Radar
Classifies replies by intent and stages drafts for your approval, so you triage and qualify the inbox against your ICP without reading every DM by hand.
Warm-by-default
Reactions, comments, and profile views before the connection ask. You show up as someone who has already engaged, not a stranger with a pitch.
Hosted state
Persistent scheduling on our infrastructure - the one thing raw Claude Code or Codex cannot hold alone. Campaigns advance and the inbox gets classified while your agent sleeps.
Install for Claude, ChatGPT, Codex, Cursor, Gemini, VS Code and any MCP client
How to add the LinkedIn MCP to Claude, ChatGPT, Codex, Cursor, Gemini, Perplexity or VS Code
Every client uses the same server, https://mcp.zevari.ai/mcp, and the same browser sign-in to your Zevari account. What changes is where you add it: a custom connector in Claude.ai, a command in Claude Code, a Developer mode connector in ChatGPT, two commands in Codex, an install link in Cursor, a custom app in Gemini, a custom connector in Perplexity, or an HTTP server in VS Code. Grok Bot in Cursor uses the same hosted endpoint as any other MCP client. Clients that cannot open a browser can use an MCP access token instead. Per-client install guides:
LinkedIn MCP for Claude
A custom connector in Claude.ai and Claude Desktop, or one command in Claude Code.
LinkedIn MCP for Codex
Two commands: codex mcp add, then codex mcp login to sign in.
LinkedIn MCP for ChatGPT
A Developer mode connector with one URL and a browser sign-in.
LinkedIn MCP for Cursor
One install link or one mcp.json entry, then a browser sign-in.
LinkedIn MCP for Gemini
Custom apps in the Gemini app, Gemini CLI and Antigravity.
LinkedIn MCP for VS Code
MCP: Add Server in VS Code for GitHub Copilot.
LinkedIn MCP for Claude Cowork
One Claude connector, then turn Zevari on in a Cowork task.
LinkedIn MCP for Claude Duet
One Claude connector, then Duet runs your LinkedIn outbound from your phone.
LinkedIn MCP for Perplexity
A remote custom connector with OAuth.
LinkedIn MCP prompts
Copy-paste prompts, one for each specialist.
LinkedIn connection limits
What LinkedIn publishes about invitation limits, and the ceilings Zevari enforces.
MoltSets for LinkedIn
Use verified contact data as input, then let Zevari handle LinkedIn enrichment and approval-gated action.
Which clients work with the Zevari LinkedIn MCP?
Any MCP client. The endpoint and sign-in are the same for all of them; Claude, Claude Cowork, Claude Duet, ChatGPT, Codex, Cursor, Gemini, Perplexity and VS Code have their own guides because the setup step differs.
| Client | How it connects | Setup page |
|---|---|---|
| Claude.ai and Claude Desktop | Customize → Connectors → Add custom connector with https://mcp.zevari.ai/mcp, then approve in the browser | Guide |
| Claude Code | claude mcp add --transport http zevari https://mcp.zevari.ai/mcp, then /mcp to sign in | Guide |
| Codex | codex mcp add zevari --url https://mcp.zevari.ai/mcp, then codex mcp login zevari | Guide |
| ChatGPT | Developer mode connector with the same URL, then approve in the browser | Guide |
| Claude Cowork | The same custom connector as Claude.ai, then turn Zevari on in a Cowork task | Guide |
| Claude Duet | The same custom connector as Claude.ai; Duet uses the connectors on your Claude account | Guide |
| Cursor, including Grok Bot in Cursor | One install link, or https://mcp.zevari.ai/mcp in ~/.cursor/mcp.json. Grok Bot in Cursor is an MCP client, not xAI Grok; nothing Grok-specific to install | Guide |
| Gemini app, Gemini CLI and Antigravity | Custom app in the Gemini app, gemini mcp add --transport http with the same URL, or serverUrl in Antigravity's mcp_config.json | Guide |
| Perplexity | Remote custom connector with OAuth and Streamable HTTP | Guide |
| VS Code with GitHub Copilot | MCP: Add Server → HTTP with the same URL, then approve in the browser | Guide |
| OpenClaw, Hermes Agent | MCP server entry or REST with a workspace token | Guide |
| Your own code, cron, or script | REST API with a bearer token | Developer docs |
Official LinkedIn MCP
Does LinkedIn have an official MCP server?
No. LinkedIn has not published an MCP server, and its developer documentation describes none. LinkedIn's own app inside ChatGPT is a tool for looking up professionals. Every LinkedIn MCP server you can install today is built by a third party: open-source servers you run on your own machine, many of which reuse a browser signed in to your LinkedIn account, or hosted servers such as Zevari, which connects through a managed session with no browser extension or cookie injection and enforces published limits on every account.
| Kind of LinkedIn MCP server | How it reaches LinkedIn | What to know |
|---|---|---|
| Official LinkedIn MCP | Does not exist. LinkedIn has published no MCP server. | Its own app inside ChatGPT is a people lookup, not an MCP server. |
| Open-source, run on your machine | Usually a browser signed in to your LinkedIn account, driven by your agent | Usually nothing enforces LinkedIn's limits for you, and the machine has to stay on for it to run. |
| Hosted (Zevari) | A managed session on Zevari's servers, with no browser extension or cookie injection | The Safety Center enforces weekly and daily ceilings; sign in with OAuth or an MCP access token. |
Comparing servers by name? See the best LinkedIn MCP servers. Worried about limits? Read what LinkedIn publishes about connection limits.
Is a LinkedIn MCP safe? Will it get me banned?
"I don't want to be banned" is the first thing every buyer says, and it is the right question. Many LinkedIn automation tools rely on browser extensions or your session cookies. Zevari is a hosted connector with neither. Every send waits for your approval until you switch a trained specialist to Autopilot, and the limits are enforced in code: 100 connection requests a week on Free*, 150 on Premium, 200 on Sales Navigator, a daily cap on all actions, burst limits, working hours and duplicate checks. If Zevari is the only tool running outbound on your LinkedIn account and you use the account lightly yourself while it works, we guarantee it; if another automation tool runs on the same account, or you use it heavily by hand at the same time, we cannot.
* Free accounts: LinkedIn caps connection requests that carry a note at 3 a month. The weekly ceiling covers requests sent without a note, and Premium and Sales Navigator accounts get a far larger allowance for requests with a note. Most outreach uses a note, so Premium or Sales Navigator is the practical choice for real outbound - the full explanation is on /safety.
FAQ
LinkedIn MCP questions
What is a LinkedIn MCP?
MCP (Model Context Protocol) is the open standard that gives an agent hands. A LinkedIn MCP exposes LinkedIn as tools an agent can call: search profiles, score a prospect, draft a connection note, send a message, comment on a post, classify an inbox reply. Zevari is the hosted LinkedIn MCP. The endpoint is free to add; Operator is the platform subscription. Every write is staged for your approval until you switch a trained specialist to Autopilot. Connect Claude Code, Codex, ChatGPT, Grok Bot in Cursor, or any MCP client - or call the same engine over REST.
Does LinkedIn have an official MCP server?
No. LinkedIn has not published an MCP server, and its developer documentation describes none. LinkedIn's own app inside ChatGPT is a tool for looking up professionals. Every LinkedIn MCP server you can install today is built by a third party: open-source servers you run on your own machine, many of which reuse a browser signed in to your LinkedIn account, or hosted servers such as Zevari, which connects through a managed session with no browser extension or cookie injection and enforces published limits on every account.
What is the difference between the MCP and the REST API?
Same engine, two front doors. Connect over MCP from Claude, ChatGPT, Codex, or any MCP client and your AI chooses from the 191 MCP tools for you. Or call the REST API from your own code when your code decides the order of calls. Both run the same safety model and approval gate.
Will this get my LinkedIn account banned?
No. If Zevari is the only tool running outbound on your LinkedIn account, and you use the account lightly yourself while Zevari is working, we guarantee it. Zevari's Safety Center paces every action like a person and counts every action the account takes against its limits: 100 connection requests a week on Free* accounts, 150 on Premium and 200 on Sales Navigator, daily caps, working hours and burst limits. Anything that would cross a limit waits for the window to reset. If another automation tool runs on the same account, or you use it heavily by hand at the same time, those actions fall outside Zevari's count, and we cannot guarantee it. The weekly figure covers connection requests sent without a note: LinkedIn caps requests that carry a note at 3 a month on a free account, and gives Premium and Sales Navigator a far larger allowance, so Premium or Sales Navigator is the practical choice for real outbound (see /safety).
How many connection requests and messages can I send per week?
Zevari enforces the limit for you: 100 connection requests a week on Free*, 150 on Premium, 200 on Sales Navigator, plus a daily cap on all actions. You do not need Sales Navigator to start; it only raises the weekly limit. Free accounts: LinkedIn caps connection requests that carry a note at 3 a month. The weekly ceiling covers requests sent without a note, and Premium and Sales Navigator accounts get a far larger allowance for requests with a note. Most outreach uses a note, so Premium or Sales Navigator is the practical choice for real outbound - the full explanation is on /safety.
Is there a warm-up sequence before outreach?
Yes. Zevari is warm-by-default - it can react to, comment on, and view a prospect's profile before sending the connection ask, which raises accept rates and lowers risk.
Which clients can I run Zevari in?
Claude.ai, Claude Desktop, Claude Code, Claude Cowork, Claude Duet, ChatGPT, Codex, Cursor (including Grok Bot in Cursor), the Gemini app, Gemini CLI and Antigravity, Perplexity, GitHub Copilot in VS Code, and any other MCP client. OpenClaw, Hermes Agent and your own code can use MCP or the REST API. Every client gets the same tools and the same Operator subscription; only the setup step differs.
Can I run LinkedIn outbound via Grok Bot?
Yes. Grok Bot in Cursor is an MCP client, not xAI Grok. Connect it to the same hosted Zevari endpoint (or REST). Same approval gates: every send is staged for you to approve until you switch a trained specialist to Autopilot. Same Operator subscription. Not a Grok-only product.
Do I need to use the same email for Claude and Zevari?
No. Your AI client account and your Zevari account are separate and can use different emails. The connection is granted to the Zevari account you sign in with.
Can it post content, and do I approve it first?
Yes. Zevari drafts content in your Voice DNA and posts only after you approve it, until you switch a trained specialist to Autopilot - the same as every other write action.
Is Zevari affiliated with LinkedIn?
No. Zevari is independent and is not affiliated with, endorsed by, or sponsored by LinkedIn Corporation. LinkedIn is a trademark of LinkedIn Corporation.
Can I connect Zevari's MCP with an access token instead of an OAuth sign-in?
Yes. Besides the browser sign-in, Zevari offers MCP access tokens, which you create under Settings → Developers. A token needs no OAuth sign-in, so it suits automations, agents and servers that cannot open a browser. Each token is locked to one LinkedIn sender account and to one workspace (or, if you confirm it, to all your workspaces), expires after 30, 90 or 365 days, and can be revoked at any time. Agencies and teams with several workspaces or senders make one token per client and sender, so an automation never has to switch workspace or sender mid-run.
A full outbound department, running today.
Your AI SDR team for LinkedIn and email, from the AI client you already use. Operator is $87/mo, or $497/yr (save 52%), and keeps campaigns running when your chat is closed.
Connect to Claude Code
Add Zevari to Claude Code with one command, or to Claude.ai as a custom connector. Sign in once and the six specialists are ready, with every send waiting for your approval until you switch a trained specialist to Autopilot.
Connect your agent
ChatGPT, Codex, Grok Bot in Cursor, OpenClaw, Hermes Agent, or your own code. Same hosted server over MCP or REST, and sends wait for your approval until you switch a trained specialist to Autopilot.
Zevari - the LinkedIn MCP with a full outbound department behind it. Your agent sleeps; your pipeline doesn't.