Connect your AI assistant to the AIARCO MCP Server in under 2 minutes. No installation required — the server is fully hosted.

​
Prerequisites

  • A channel API key from publishers.ads.aiarco.com (starts with aiarco_)
  • An MCP-compatible AI client — Claude Code, VS Code Copilot, Cursor, Windsurf, or similar

​
Server URL

https://mcp.aiarco.com/mcp
This is the only endpoint you need. It supports Streamable HTTP transport with JSON-RPC 2.0 messaging.

​
Connect Your Client

​
Claude Code

Run this command in your terminal: 
claude mcp add --transport http aiarco https://mcp.aiarco.com/mcp

​
VS Code (Copilot)

Add to your .vscode/mcp.json: 
{
  "servers": {
    "aiarco": {
      "type": "http",
      "url": "https://mcp.aiarco.com/mcp"
    }
  }
}

​
Cursor / Windsurf

Add to your MCP configuration file (.cursor/mcp.json or equivalent): 
{
  "mcpServers": {
    "aiarco": {
      "url": "https://mcp.aiarco.com/mcp"
    }
  }
}

​
Any MCP Client

Use these connection details for any MCP-compatible client: 
Endpoint:   https://mcp.aiarco.com/mcp
Transport:  Streamable HTTP
Format:     JSON-RPC 2.0

​
Connection Lifecycle

  1. Initialize — Your client sends an initialize request to negotiate capabilities
  2. Discover — The server responds with available tools via tools/list
  3. Invoke — Your client calls tools like broadcast_signal and get_offers as needed
  4. Track — Report impressions with track_impressions after displaying offers

​
Verify Connection

Test your connection with the MCP Inspector: 
npx -y @modelcontextprotocol/inspector
Enter https://mcp.aiarco.com/mcp as the endpoint. You should see the available tools listed.

​
Next Steps

Tool Reference

Explore parameters, schemas, and examples for each tool.

Overview

Learn how signal collection and offer delivery work.