Foxcite MCP Connector

A transparent Model Context Protocol (MCP) proxy that natively connects Claude Desktop (and other MCP-compatible clients) directly to your Foxcite AEO data.

With this connector, you can leverage ai search visibility for agencies by asking Claude to "analyze my brand's AI visibility," "check my latest citation gaps," or "run a new audit for [query]"—and Claude will seamlessly interface with your Foxcite workspace.

Installation & Configuration

The connector is designed to be run as an npx executable directly from your Claude Desktop configuration file.

1. Get an API Key

Generate an API Key from your Foxcite Dashboard.

2. Configure Claude Desktop

Open your Claude Desktop configuration file.

• Mac: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

Add the Foxcite MCP server to your mcpServers object:

{
  "mcpServers": {
    "foxcite": {
      "command": "npx",
      "args": ["-y", "@foxcite/mcp-connector"],
      "env": {
        "FOXCITE_API_KEY": "fc_live_..."
      }
    }
  }
}

3. Restart Claude

Fully quit and restart Claude Desktop. You will now see the "Foxcite" tools available in the Claude interface!

Exposed MCP Tools

Once connected, the LLM gains access to the following tools:

• list_brands: Retrieve all brands and workspaces in the user's account.
• get_visibility_score: Fetch the current overall AI Visibility Score (AVS) for a specific brand.
• run_quick_audit: Trigger a live scrape against ChatGPT, Claude, Perplexity, etc. to see if a brand is recommended for a given query.
• get_citation_gaps: Retrieve specific LLM responses where competitors were cited but the user's brand was omitted.

Developer Notes

This connector acts as an SSE (Server-Sent Events) bridge to Foxcite's backend. It uses the official @modelcontextprotocol/sdk to expose Foxcite's REST API as an MCP standard interface.

Resources

• Homepage
• GitHub Repository
• Model Context Protocol Docs