Skip to main content
Beta Feature: MCP integration is currently in beta and not available to all users yet. Interested in getting access? Contact us to join the beta program.

What is MCP?

The Model Context Protocol (MCP) is an open standard that enables AI assistants like Claude, Cursor, and others to interact directly with external services. With Affonso’s MCP integration, you can manage your affiliate program using natural language commands.
MCP allows AI assistants to read and modify your affiliate data in real-time, making it perfect for quick queries, bulk operations, and automation workflows.

Why Use MCP?

Natural Language

Ask questions like “Show me my top 5 affiliates by revenue” or “Create a new affiliate for [email protected]

Real-Time Data

Get instant access to your affiliate data without switching between tools

Bulk Operations

Perform complex operations across multiple affiliates, referrals, or commissions

AI-Powered Analysis

Let AI analyze your affiliate performance and provide insights

Available Tools

Affonso MCP provides 18 tools across 4 domains:
DomainToolsDescription
Affiliates5List, get, create, update, delete affiliates
Referrals5Track clicks, leads, and conversions
Commissions5Manage commission transactions
Payouts3Process and track affiliate payouts

How It Works

  1. You ask your AI assistant a question about your affiliates
  2. The AI calls the appropriate Affonso MCP tool
  3. Affonso authenticates and executes the request
  4. Results are returned in a readable format

MCP Endpoint

https://api.affonso.io/mcp
The MCP server uses Streamable HTTP transport, which is supported by Claude Desktop, Cursor, and other MCP-compatible clients.

Authentication

MCP uses the same API keys as the REST API. You’ll need a valid API key with appropriate permissions:
  • Production: sk_live_ + 32 characters
  • Development: sk_dev_ + 32 characters
Your API key has full access to your affiliate data. Never share it publicly or commit it to version control.

Response Formats

All MCP tools support two response formats:
FormatBest For
Markdown (default)Human-readable output with formatting
JSONProgrammatic processing and detailed data
Specify the format using the responseFormat parameter on any tool.

Next Steps