DEV Community

roykeano941-eng
roykeano941-eng

Posted on

Building an MCP-Enabled Sports Data API: Making Sports Intelligence Agent-Discoverable

The Problem

AI agents are getting good at business intelligence tasks. But they still can't access most domain-specific data — it's locked in PDFs, behind paywalls, or formatted for humans.

I built a sports sponsorship intelligence API that's natively accessible to AI agents via Model Context Protocol (MCP).

What It Does

Query: GET /api/match-score?club=everton&brand=Revolut

Response: Structured JSON with:

  • Match score (1-100)
  • Reasoning
  • Comparable deals
  • Risk factors
  • Suggested deal terms

Why MCP?

MCP lets AI agents discover and use your API without custom integration. Any MCP-compatible agent finds the endpoint at:

https://sportsignal.agency/.well-known/mcp.json

The Business Case

£83.9M in Premier League sponsorship revenue is changing hands due to the UK gambling ad ban. 11 clubs need new sponsors. I built the scoring layer for this market.

Scoring Algorithm

  • Fan base reach: 30%
  • Digital presence: 20%
  • Category fit: 35%
  • Sponsorship gap urgency: 15%

Try It

Free tier: all 11 gambling-affected PL clubs → sportsignal.agency/match-score

Feedback welcome — especially on MCP implementation patterns.

Top comments (0)