About LinkedNav
LinkedNav MCP connects AI assistants to a LinkedNav account for running LinkedIn outreach from chat. Its tools cover lead discovery, campaign management, messaging, and performance analysis.
Key features
- 01Find leads by buying signals and other criteria
- 02Create, launch, pause, and schedule campaigns
- 03Write and save connection notes and follow-ups
- 04Send invitations, read and answer messages
- 05Enrich contact and profile details
- 06Audit setup and review funnel analytics
Use cases
- 01Find target contacts who recently changed jobs
- 02Launch and schedule LinkedIn outreach campaigns
- 03Review replies and prepare follow-up responses
- 04Analyze an outreach funnel and check setup
MCP support
It provides a hosted MCP server at https://mcp.linkednav.com/, exposing LinkedNav account tools to MCP clients such as Claude and Cursor. Connection uses browser OAuth or an API key.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.linkednav.com/claude mcp add --transport http linglistack-linkednav https://mcp.linkednav.com/{"mcpServers":{"linglistack-linkednav":{"type":"http","url":"https://mcp.linkednav.com/"}}}Open in Cursor{"servers":{"linglistack-linkednav":{"type":"http","url":"https://mcp.linkednav.com/"}}}Open in VS Code{"mcpServers":{"linglistack-linkednav":{"type":"http","url":"https://mcp.linkednav.com/"}}}{"mcpServers":{"linglistack-linkednav":{"serverUrl":"https://mcp.linkednav.com/"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
https://mcp.linkednav.com/Auth requiredTools
Details
From the project READMELinkedNav MCP
Run your LinkedIn outreach from your AI chat.
LinkedNav MCP connects AI assistants (Claude, Cursor, and any other MCP app) to your LinkedNav account. Once it's connected, you just ask your assistant to find the right people, launch campaigns, send connection requests and messages, answer replies, and check results. It does the outreach for you.
- Server:
https://mcp.linkednav.com/ - Works with: Claude Code, Claude Desktop, claude.ai, Cursor, and any MCP client
- Sign up (free trial): linkednav.com
What you can do with it
Ask your assistant things like:
- "Who should I target on LinkedIn, and what should I say?"
- "Find people who just changed jobs at companies like these."
- "Set up a campaign and turn it on."
- "What replies came in, and what should I say back?"
- "How is my outreach doing, and what should I fix?"
Under the hood it gives your assistant tools for:
- Setup and targeting (your business and your ideal customer)
- Finding leads (buying-signal search, competitor and influencer engagers, social listening)
- Campaigns (create, launch, pause, schedule)
- Messages (write and save your connection note and follow-ups)
- Sending and inbox (send connects and messages, read and answer replies)
- Enrichment (find emails, phones, and profile details)
- Analytics (audit your setup, see your funnel, find what to fix)
Connect it (about 2 minutes)
- Sign up at linkednav.com and create an API key at /app/api-keys.
- Add the server to your app.
Claude Code:
claude mcp add --transport http linkednav https://mcp.linkednav.com/ \
--header "Authorization: Bearer YOUR_KEY"
Or install the plugin in this repo, which signs in through your browser. See The plugin.
Claude Desktop or claude.ai: Settings, then Connectors, then Add custom connector.
Name it LinkedNav, URL https://mcp.linkednav.com/, then sign in with your browser
(OAuth) or add an Authorization: Bearer YOUR_KEY header.
Cursor / Cline / other MCP apps (~/.cursor/mcp.json):
{
"mcpServers": {
"linkednav": {
"url": "https://mcp.linkednav.com/",
"headers": { "Authorization": "Bearer YOUR_KEY" }
}
}
}
The plugin
This repo is also a plugin for Claude Code. It holds two things:
- The MCP server
linkednav, a pointer tohttps://mcp.linkednav.com/(.mcp.json). Nothing is downloaded or run on your machine. - The skill
linkednav(skills/linkednav/SKILL.md): how to use the tools well. It tells your assistant to ask you before anything is sent or changed on LinkedIn, and to report what the tools return instead of guessing.
To install it, run these in a terminal:
claude plugin marketplace add linglistack/linkednav-mcp
claude plugin install linkednav@linkednav
Then start Claude Code, run /mcp, choose linkednav, and sign in. The sign-in opens
LinkedNav's connect page in your browser, where you paste your API key once.
Sign-in and network
The plugin connects to https://mcp.linkednav.com and reaches nothing else.
https://mcp.linkednav.com/: the MCP server (Streamable HTTP).https://mcp.linkednav.com/.well-known/oauth-protected-resourceand/.well-known/oauth-authorization-server: where your app finds the sign-in.https://mcp.linkednav.com/register,/connectand/oauth/token: the sign-in itself, OAuth with PKCE. Your app registers itself; there is no client secret.https://mcp.linkednav.com/connect: the page where you paste your API key.
What it needs from you: a LinkedNav account and an API key from /app/api-keys. No key is stored in the plugin, and it has no hooks and no scripts.
Pairs with the open-source skill
Want a ready-made outreach team that uses this MCP? Install the free, open-source LinkedIn Outreach OS skill. The skill writes your plan and messages; this MCP does the real work.
About this repo
This repo holds the public info and the MCP registry manifest (server.json) for the
LinkedNav MCP server. The server itself is hosted at mcp.linkednav.com and runs on your
own LinkedNav account. The server's source code is not part of this repo.
License
MIT for the contents of this repo. LinkedNav is a hosted service.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.