linkmcp
Hosted LinkedIn MCP server. Connect your own LinkedIn account to Claude, ChatGPT, Codex, Cursor, Gemini CLI or any MCP client: 34 tools for search, profiles, inbox, posts, company pages and email finding. 7-day free trial.
Install / Use
claude mcp add linkmcp-io -- npx -y github:linkmcp-io/linkmcpIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
AutomationSupported Platforms
Tags
Our assessment of linkmcp
linkmcp scores 75/100 on our quality scale, 2608th of 2,889 Automation skills we index.
Its MCP Server is 11 KB long, well organised into 16 sections with 16 code examples: a thorough specification that gives an agent plenty to work with.
It has 3 GitHub stars, so there is little community track record yet; judge it on its content.
Maintenance, license and trust
- The repository was last updated 4 days ago, so linkmcp is actively maintained.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 92/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
linkmcp compared with similar skills
All 4 of these similar skills score higher than linkmcp; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| linkmcp (this skill)by linkmcp-io | 75 | 3 | 4d ago | MCP Server |
| Agent-Reachby Panniantong | 100 | 95.0k | 1d ago | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 74.9k | today | CLAUDE.md |
| CowAgentby zhayujie | 100 | 47.3k | today | CLAUDE.md |
| Scraplingby D4Vinci | 100 | 86.5k | 1d ago | MCP Server |
Frequently asked questions
- How do I install linkmcp?
- Run
claude mcp add linkmcp-io -- npx -y github:linkmcp-io/linkmcp. The install tabs above show the steps for each supported agent. - Which AI agents does linkmcp work with?
- It is written for Claude Code, Claude Desktop, Cursor, Gemini CLI and OpenAI Codex, as a MCP Server file. Other agents that read the same format can often use it too.
- Is linkmcp safe to use?
- It is MIT-licensed and scores 92/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is linkmcp still maintained?
- The repository was last updated 4 days ago, so linkmcp is actively maintained.
Skill content
View source on GitHubLinkMCP - hosted LinkedIn MCP server
Your own LinkedIn account, as tools for Claude, ChatGPT, Codex, Cursor and Gemini.
LinkMCP is a hosted MCP server for LinkedIn. You connect your LinkedIn account once in the web app. Then your AI assistant can research people and companies, search LinkedIn (also Sales Navigator, with your own seat), read and answer your LinkedIn messages, post, comment and react (as yourself or as a company page that you administer), manage invitations, and find work emails and mobile numbers. There is nothing to install and nothing to run.
No card. Sign in with an email code or Google. Setup takes about a minute.
Quick start
- Start the trial at app.linkmcp.io.
- Add LinkMCP to your AI client with the steps for your client below. Your client opens a browser window. Sign in to LinkMCP and approve.
- Ask. Copy one of the example prompts.
The endpoint is the same for every client:
https://app.linkmcp.io/api/mcp
Transport: Streamable HTTP. Sign-in: OAuth in the browser. If your client cannot do OAuth, send an access key as Authorization: Bearer <key>. You create the key in the web app under API Keys.
What works in the free trial: profile and company lookups, the posts and comments of people and companies, and email and phone finding. To act as yourself on LinkedIn (people search, inbox, invitations, posting, company pages), subscribe and connect your LinkedIn account once in the web app.
Claude (web and desktop)
Add to Claude opens the "Add custom connector" dialog with the name and URL filled in. Or do it by hand:
- In Claude, open Customize > Connectors, click + and choose Add custom connector.
- Name it
LinkMCPand pastehttps://app.linkmcp.io/api/mcp. - Click Add, then Connect, and approve the sign-in window.
On Team and Enterprise plans, an Owner adds the connector in Organization settings > Connectors. Then each member clicks Connect. The Free plan allows one custom connector.
Claude Code
claude mcp add --transport http linkmcp https://app.linkmcp.io/api/mcp
Then run /mcp in Claude Code, select linkmcp and choose Authenticate.
ChatGPT
Developer mode needs ChatGPT Plus, Pro, Business, Enterprise or Edu (not Free), on the web.
- Open Settings > Security and login and turn on Developer mode.
- Go to chatgpt.com/plugins, click +, name it
LinkMCPand pastehttps://app.linkmcp.io/api/mcp. - Create the connection and approve the sign-in window.
- In a new chat, click + > Developer mode and turn on LinkMCP.
A workspace admin can turn Developer mode off for a Business or Enterprise workspace.
Codex
Codex CLI:
codex mcp add linkmcp --url https://app.linkmcp.io/api/mcp
Codex opens the sign-in window. If it does not, run codex mcp login linkmcp.
Codex app and IDE extension: open the settings menu, select MCP servers > Add server, name it LinkMCP, choose Streamable HTTP, paste the endpoint, save, and click Authenticate.
Cursor
Or add this to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (one project):
{
"mcpServers": {
"linkmcp": { "url": "https://app.linkmcp.io/api/mcp" }
}
}
Restart Cursor and approve the sign-in window.
Gemini CLI
Install LinkMCP as a Gemini CLI extension:
gemini extensions install https://github.com/linkmcp-io/linkmcp
Or add only the server:
gemini mcp add --transport http linkmcp https://app.linkmcp.io/api/mcp
Then start Gemini CLI and run /mcp auth linkmcp to sign in.
Other clients (VS Code, Windsurf, n8n, Make, Zapier)
VS Code:
code --add-mcp '{"name":"linkmcp","type":"http","url":"https://app.linkmcp.io/api/mcp"}'
- Windsurf and other MCP clients: add a remote MCP server with the endpoint above.
- n8n, Make, Zapier: use the MCP client node or module of the platform with the endpoint and an access key. Step by step: n8n guide, Make guide.
Example prompts
Copy a prompt, replace the parts in angle brackets, and paste it into your AI client.
These work in the free trial:
Look up the company linkedin.com/company/stripe on LinkedIn and give me a short
brief: what they do, how big they are and what they posted recently.
Look up the LinkedIn profile <profile URL> and give me a short brief: role, company,
time in the role, and what they posted in the last month.
Find the work email of <full name> at <company domain> and validate it.
Here are LinkedIn profile URLs: <URLs>. Make a table with name, title, company and
location.
These need a paid plan, with your LinkedIn account connected:
Find heads of RevOps at Series B SaaS companies in the Netherlands. Summarise what
each one posted in the last month.
List my LinkedIn conversations with no reply in 7 days. Check what each person
posted recently and draft a reply for each. Show me the drafts before you send
anything.
Who do I know at <company>? Show my shared connections with their VP of Sales and
draft a short intro request.
Who reacted to my last post? Which of them are marketing directors that I am not
connected to? Draft a connection note for the top 10.
List the LinkedIn company pages that I administer. Draft a post for our page about
<topic>. Show it to me, and publish it as the page only after I say OK.
The 34 tools
Tools marked trial work in the free trial. All tools work on every paid plan.
| Category | Tools |
|---|---|
| Profiles and companies | linkedin_get_profile (trial), linkedin_get_company (trial, by URL or ID), linkedin_bulk_get_profiles (trial), linkedin_bulk_get_companies (trial) |
| Search | linkedin_search_people, linkedin_search_sales_navigator, linkedin_search_jobs |
| Posts and engagement | linkedin_get_person_posts (trial), linkedin_get_company_posts (trial), linkedin_get_post_comments (trial), linkedin_get_nested_comments, linkedin_get_post_reactions |
| Messaging | linkedin_list_conversations, linkedin_get_conversation_messages, linkedin_send_message (also InMail), linkedin_mark_conversation_read |
| Connections | linkedin_get_connections, linkedin_get_shared_connections, linkedin_list_connection_requests, linkedin_manage_connection_request, linkedin_send_connection_request |
| Content | linkedin_create_post, linkedin_comment_on_post, linkedin_react_to_post |
| Company pages | linkedin_list_company_pages, plus the as_page option on linkedin_create_post, linkedin_comment_on_post and linkedin_react_to_post |
| Your LinkedIn | linkedin_who_am_i, linkedin_get_post_analytics, linkedin_get_profile_views, linkedin_get_my_engagement, linkedin_get_my_saved_items |
| Contact finding | find_email (trial), validate_email (trial), find_mobile (trial) |
| Utility | send_feedback |
26 tools only read. 8 tools write: send a message, send a connection request, accept, decline or withdraw a connection request, create a post, comment, react, mark a conversation as read, and send feedback. Every tool has a title and a read-only or write annotation, so your client can ask you before a write. Full parameter documentation: llms-full.txt.
Company pages: your AI can post, comment and react as a LinkedIn company page that you administer. It must name the page with as_page. Without as_page, it acts as you. Actions as a page count toward the same daily limits as your own actions.
Pricing
- Free trial: 7 days, no card. The tools marked trial work at once. Connecting your own LinkedIn account needs a paid plan.
- Starter: $19 per month. If you subscribe from a new trial, your first month is $9.50.
- Pro: $49 per month (5x the Starter usage). Max: $199 per month (50x the Starter usage).
- Every paid plan includes every tool. Each plan has a monthly usage allowance and a hard cap, so a looping agent cannot run up a bill. Top-up packs cost $10.
Current prices: app.linkmcp.io/#pricing
Account risk and limits
Any automated use of LinkedIn can put an account at risk, and no tool can promise that LinkedIn will not restrict an account. LinkMCP limits the risk:
- Server-side rate limits on every call, and a warm-up limit for connection requests on new or quiet accounts.
- Your LinkedIn password is not stored. The LinkedIn session runs on dedicated session infrastructure, not in your browser.
- No bulk scraping and no mass exports. No training on your data.
- Read Will my LinkedIn get banned? before you automate at volume.
FAQ
Is LinkMCP open source? No. This repository holds the documentation and client configuration. The server is a hosted service, and its code is not public. There is nothing to install or run locally.
Is LinkMCP an official LinkedIn product? No. LinkMCP is an independent product of Third Person Systems LLC. It is not affiliated with or endorsed by LinkedIn, and it does not use an official LinkedIn partner API.
Do I need Sales Navigator? No. LinkMCP works with a normal LinkedIn account. If you have a Sales Navigator seat, the Sales Navigator search tool uses it.
Which AI clients work? Every client that supports remote MCP servers: Claude (web, desktop, Claude Code), ChatGPT (Developer mode), Codex, Cursor, Gemini CLI, VS Code, Windsurf, n8n, Make and Zapier.
Guides
- Connect LinkedIn to ChatGPT
- Connect LinkedIn to Claude
- Connect LinkedIn to n8n
- Connect LinkedIn to Make
- LinkedIn outbound funnel recipe for Make and n8n
- LinkMCP and other LinkedIn MCP options
- LinkedIn MCP vs Expandi and PhantomBuster
- Will my LinkedIn get banned?
- More guides: app.linkmcp.io/guides. Docs for LLMs: llms.txt
Support
- In your AI client, ask it to use the
send_feedbacktool. - GitHub issues for documentation and configuration problems.
- Email: info@linkmcp.io
© Third Person Systems LLC. LinkMCP is independent. It is not affiliated with, endorsed by, or built on an official API of LinkedIn.
Related Skills
Agent-Reach
95.0kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
headroom
74.9kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
CowAgent
47.3kOpen-source personal AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, multi-channel. Lightweight, extensible, one-line install.
Scrapling
86.5k🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ and follow here for daily tips and tricks: https://x.com/Scrapling_dev
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
