Hyros MCP: Connect Hyros to Any AI Agent or Automation

automation
Hyros MCP: Connect Hyros to Any AI Agent or Automation

If you run paid ads and use Hyros for attribution, you've probably hit the same wall: your attribution data lives inside Hyros. You can see it in the dashboard, pull reports manually, copy-paste numbers into spreadsheets — but you can't feed it into AI workflows, automation pipelines, or custom dashboards without wrestling the API yourself.

That's the exact problem hyros-mcp solves. It's the first open-source Model Context Protocol (MCP) server for Hyros, built by Carlos Aragón (GitHub, npm) — founder of Vixi Agency and a Hyros OG member since 2020.

It hit 100+ downloads in the first 24 hours. That tells you how many people were waiting for this.

What Is the Model Context Protocol?

Model Context Protocol (MCP) is an open standard created by Anthropic. It defines how AI agents — Claude, ChatGPT, and other LLMs — connect to external data sources and tools in a structured, standardized way.

Before MCP, every integration was bespoke: you'd write a custom API wrapper, wire it into your agent framework, handle authentication yourself, deal with pagination and rate limits. Every tool, every agent, every integration was built from scratch.

MCP changes that. It's the USB-C equivalent for AI integrations — one standard protocol, works everywhere. Any MCP server can plug into any MCP-compatible client. That means:

  • Build the Hyros MCP server once
  • Use it with Claude Desktop, ChatGPT, Cursor, or any agent framework
  • No custom wiring per client or use case

MCP is supported natively in Claude and increasingly across the AI ecosystem. It's where agent infrastructure is heading.

What hyros-mcp Gives You

Once you install and configure hyros-mcp, your AI agent gains live access to your Hyros account. Specifically, it can:

  • Pull campaign performance — ROAS, revenue, spend, conversions by source
  • Query lead attribution — see the full journey from ad click to sale for any lead
  • Get conversion events — purchases, trials, subscriptions with timestamps and sources
  • Analyze ad source performance — compare Facebook vs. Google vs. organic vs. email attribution
  • Answer attribution questions in natural language — "which campaign drove the most revenue last week?" returns real data, not a hallucination

This transforms Hyros from a dashboard you check into a data source your AI agents actively reason about.

Real Use Cases for Hyros AI Agent Integration

Automated Client Attribution Reports

You spend hours every Friday pulling Hyros data, formatting it, and sending it to clients. With the Hyros MCP + an n8n or Claude workflow, an AI agent can query Hyros directly, structure the data, write narrative analysis, and deliver the report automatically — on schedule, without human touchpoints.

ROAS-Triggered Campaign Alerts

A workflow checks your Hyros ROAS by campaign every morning. If any campaign drops below your target threshold, the AI agent flags it in Slack with the attribution data attached, and optionally triggers a budget pause via the ad platform API.

AI-Powered Hyros Dashboard

Build a Claude-backed dashboard where your team (or your clients) asks attribution questions in plain English. "Which ad source has the best 90-day LTV?" The agent queries Hyros live and responds with data — not a placeholder.

Hyros Claude Integration for Agency Ops

For agencies managing multiple Hyros accounts, an AI agent with MCP access can aggregate attribution across all client accounts, generate cross-account performance summaries, and surface insights that would take hours to compile manually.

Getting Started

npm install hyros-mcp

You'll need your Hyros API key. For Claude Desktop, add this to claude_desktop_config.json:

{
  "mcpServers": {
    "hyros": {
      "command": "npx",
      "args": ["hyros-mcp"],
      "env": {
        "HYROS_API_KEY": "your_api_key_here"
      }
    }
  }
}

Restart Claude Desktop and you'll see Hyros tools available. Ask: "What were my top performing campaigns this month?" — Claude queries your Hyros account and returns real data.

Full setup guide and documentation: github.com/CachoMX/Hyros-MCP

Why This Exists: Hyros Expertise Meets AI

I'm Carlos Aragón. I've been a Hyros OG member since 2020 — implementing Hyros for agencies, media buyers, coaching programs, and SaaS companies. I've seen every setup mistake and every breakthrough.

I also build AI agents. When I started building agents that needed to reason about attribution data, the gap was obvious. The Hyros API is solid and well-documented. MCP is the right protocol. The bridge just didn't exist yet.

I built it, tested it across our internal stack, and open-sourced it because the Hyros community deserves clean tooling.

If 100+ people downloaded it in the first 24 hours, there were a lot of agencies and builders who were waiting for exactly this.

Hyros API Integration Without the Boilerplate

Before hyros-mcp, a custom Hyros API integration looked like:

  1. Dig through API documentation
  2. Build authentication handling (API key headers, retry logic)
  3. Handle pagination for campaign and lead data
  4. Write normalization code for API responses
  5. Maintain it as Hyros updates their API

With hyros-mcp, steps 1–5 are handled. You configure your API key, and your AI agent has immediate, structured access to Hyros attribution data.

For agencies building client-specific automations, this is the difference between a week of setup and an afternoon.

Frequently Asked Questions

What is Hyros MCP? Hyros MCP is an open-source Model Context Protocol server that gives AI agents — including Claude and ChatGPT — live, structured access to Hyros attribution data via the Hyros API. Built by Carlos Aragón of Vixi Agency.

What is hyros-mcp used for? It's used to connect AI agents and automation workflows to Hyros attribution data. Common uses include automated attribution reports, ROAS monitoring, AI-powered client dashboards, and Hyros data pipelines in n8n or other automation tools.

Does hyros-mcp work with Claude? Yes. Claude supports MCP natively through Claude Desktop and the Anthropic API. Once configured, Claude can pull live campaign performance, lead attribution, and conversion data from your Hyros account.

Does hyros-mcp work with ChatGPT or other AI tools? Yes. MCP is an open protocol. Any AI agent framework supporting tool-use — OpenAI, LangChain, CrewAI, AutoGen, n8n AI nodes — can work with an MCP server. The implementation details vary by platform; see the GitHub docs for specifics.

Is hyros-mcp an official Hyros product? No. It's an open-source community package built by Vixi Agency. It uses the official Hyros API and is not affiliated with or endorsed by Hyros Inc.

How many people use hyros-mcp? The package hit 100+ downloads in its first 24 hours of release on npm. Download count continues to grow as more agencies discover it.

Where is the source code? Fully open-source at github.com/CachoMX/Hyros-MCP.

Next Steps