# LinkedNav [Health: Active]

**Category:** 🌐 Social Media  
**Repository:** https://github.com/linglistack/linkednav-mcp  
**GitHub Stars:** 0  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/linkednav

## Description
Run LinkedIn outreach from your AI chat: find leads, launch campaigns, send, and reply.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "linkednav": {
    "url": "https://mcp.linkednav.com/"
  }
}
```

## Documentation

## What LinkedNav does

LinkedNav MCP server brings LinkedIn outreach operations into MCP-compatible AI assistants through a hosted endpoint at `https://mcp.linkednav.com/`. It operates against the user's LinkedNav account rather than running the service locally. After authentication, an assistant can help define targeting criteria, identify prospects, create outreach campaigns, draft messages, send connection requests and follow-ups, read replies, and inspect campaign results.

The server covers both execution and review. Lead discovery includes buying-signal searches, competitor and influencer engagers, and social listening. Enrichment can return email addresses, phone numbers, and profile details. Analytics tools support setup audits, funnel inspection, and identifying areas to improve.

## How it works

The MCP connection uses Streamable HTTP. Clients can authenticate with a LinkedNav API key sent as an `Authorization: Bearer` header, or use the browser-based OAuth flow available in clients that support the connector setup. The service handles requests through the user's LinkedNav account; the repository does not contain the hosted server implementation.

Claude Code can connect directly with its MCP command or through the repository's plugin. The plugin points to the hosted server and includes a skill that instructs Claude Code to request confirmation before sending or changing anything on LinkedIn. It does not download or run the server locally.

## Setup and configuration

Create a LinkedNav account and generate an API key from the account's API keys page. Add `https://mcp.linkednav.com/` as an MCP server in the chosen client, passing the key as a bearer token where required.

Claude Desktop and claude.ai can add the endpoint as a custom connector and authenticate with a browser or bearer header. Cursor, Cline, and similar clients can configure the endpoint and header in their MCP settings. Claude Code also supports the repository plugin:

```bash
claude plugin marketplace add linglistack/linkednav-mcp
claude plugin install linkednav@linkednav
```

The plugin's sign-in uses OAuth with PKCE. It has no client secret, hooks, or scripts, and does not store the API key. Its documented network destinations are the LinkedNav MCP endpoint and its OAuth registration, authorization, token, and connection paths.

## Tools and capabilities

- Define business information and ideal-customer targeting criteria.
- Search for leads using buying signals, competitor or influencer engagement, and social listening.
- Create, launch, pause, and schedule campaigns.
- Draft and save connection notes and follow-up messages.
- Send connection requests and messages, then read and answer replies.
- Enrich leads with email, phone, and profile information.
- Audit outreach setup, inspect the funnel, and review performance.

## Limitations and notes

LinkedNav is a hosted service, while the repository contains public information and the MCP registry manifest. The repository's MIT license applies to its contents; LinkedNav itself is a hosted service. The separate LinkedIn Outreach OS project can provide planning and message-writing guidance, while LinkedNav MCP performs the outreach actions.

_Full upstream README: https://allmcps.com/mcp/linkednav/readme_