Cloudbet Sports MCP Server
Query Cloudbet sports data via AI using the Model Context Protocol.
The Cloudbet Sports MCP Server exposes structured sports betting data to AI models using Model Context Protocol, enabling natural-language queries mapped to live market data.
Why MCP?
- Query sports data using plain English
- Receive structured, real-time odds
- Compatible with Claude, OpenAI, and custom AI runtimes
- Open-source, extensible, easy to self-host
How It Works
The demo server currently implements a single tool via MCP JSON-RPC:
Tool: findEventsAndMarketsByCompetition
Returns upcoming events and primary markets for a fuzzy-matched competition.
Tool Metadata (tools/list):
{
"name": "findEventsAndMarketsByCompetition",
"description": "Finds upcoming events and odds using fuzzy search",
"inputSchema": {
"type": "object",
"properties": {
"competitionName": { "type": "string" },
"limit": { "type": "integer" }
},
"required": ["competitionName"]
}
}
Sample Call (tools/call):
{
"params": {
"name": "findEventsAndMarketsByCompetition",
"arguments": {
"competitionName": "Premier League",
"limit": 1
}
}
}
Sample Response (truncated):
{
"structuredContent": [
{
"Name": "Liverpool vs. AFC Bournemouth",
"CutoffTime": "2025-08-15T19:00:00Z",
"markets": {
"soccer.match_odds": {
"submarkets": {
"period=ft": {
"selections": [
{
"Outcome": "home",
"Price": 1.346,
"MaxStake": 574.15,
"MarketURL": "soccer.match_odds/home"
},
{
"Outcome": "draw",
"Price": 5.238,
"MaxStake": 208.02,
"MarketURL": "soccer.match_odds/draw"
},
{
"Outcome": "away",
"Price": 6.711,
"MaxStake": 208.02,
"MarketURL": "soccer.match_odds/away"
}
]
}
}
},
"soccer.total_goals": { /* includes over/under, Price, MaxStake */ },
"soccer.asian_handicap": { /* includes line, sides, Price, MaxStake */ }
}
}
]
}
Supported Sports & Markets
Primary markets selected per sport:
| Sport | Markets |
|---|---|
| Soccer | match_odds, asian_handicap, total_goals |
| Basketball | handicap, moneyline, totals |
| Tennis | winner, game_handicap, total_games |
| Others | NFL, Cricket, MMA, Boxing, Ice Hockey |
Natural Language Mapping
Fuzzy search maps user queries to competitions using Levenshtein distance:
| Example Query | Match |
|---|---|
| “Premier League odds” | soccer-england-premier-league |
| “NBA games” | basketball-usa-nba |
| “Champions League” | soccer-uefa-champions-league |
Getting Started
Self-host
git clone https://github.com/cloudbet/sports-mcp-server
cd sports-mcp-server
go run .
Claude Integration
{
"mcpServers": {
"cloudbet-sports": {
"command": "go",
"args": ["run", "main.go"],
"cwd": "/path/to/sports-mcp-server"
}
}
}
CURL Example
curl -X POST http://localhost:8080/ -H "Content-Type: application/json" -d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "findEventsAndMarketsByCompetition",
"arguments": { "competitionName": "Premier League", "limit": 3 }
}
}'