Sports MCP Server
A unified MCP server that combines The Odds API and SportMonks Football API v3 to provide 28 tools for sports data like odds, scores, fixtures, standings, and predictions, enabling Claude to interact with a wide range of sports information.
README
šļø Sports MCP Server
A Model Context Protocol (MCP) server that unifies The Odds API and SportMonks Football API v3 into a single tool surface for Claude Desktop and Claude Code.
What This Does
| Source | Coverage | Tools |
|---|---|---|
| The Odds API | All sports ā NBA, NFL, NHL, NCAAM, NCAAF, Soccer, Tennis, MMA, Golf, MLB | 10 tools: odds, scores, events, markets, participants, historical |
| SportMonks | Football (soccer) deep data ā EPL, La Liga, Bundesliga, Ligue 1, Serie A, Champions League, etc. | 18 tools: fixtures, standings, statistics, xG, predictions, odds, livescores, squads |
Total: 28 tools available to Claude via MCP.
Quick Start
1. Install dependencies
cd sports-mcp-server
npm install
npm run build
2. Set your API keys
You'll configure these in the Claude Desktop config (step 3), but for testing:
export ODDS_API_KEY="your-odds-api-key-here"
export SPORTMONKS_TOKEN="your-sportmonks-token-here"
3. Configure Claude Desktop
Add this to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"sports-data": {
"command": "node",
"args": ["/FULL/PATH/TO/sports-mcp-server/dist/index.js"],
"env": {
"ODDS_API_KEY": "your-odds-api-key-here",
"SPORTMONKS_TOKEN": "your-sportmonks-token-here"
}
}
}
}
ā ļø Replace
/FULL/PATH/TO/with the actual absolute path to the project.
4. Restart Claude Desktop
After saving the config, restart Claude Desktop. You should see the sports tools available in the MCP tools menu.
For Claude.ai (Web / Chrome) ā ā This is what you want
The web app needs an HTTP server, not stdio. Here's how:
Step 1: Build & start the HTTP server
cd sports-mcp-server
npm install
npm run build
# Start the HTTP server with your API keys
ODDS_API_KEY=your-key-here SPORTMONKS_TOKEN=your-token-here npm run start:http
You should see:
šļø Sports MCP Server (HTTP/SSE) running on http://localhost:3456
Tools: 10 Odds API + 18 SportMonks = 28 total
Odds API key: ā
configured
SportMonks token: ā
configured
Step 2: Connect in Claude.ai
- Go to claude.ai ā Settings (bottom-left gear icon)
- Click Integrations in the left sidebar
- Click Add Custom Integration
- Set:
- Name:
Sports Data - URL:
http://localhost:3456/sse
- Name:
- Save and enable it
That's it ā the 28 tools are now available in your Claude.ai conversations.
Optional: Protect with auth token
ODDS_API_KEY=xxx SPORTMONKS_TOKEN=xxx MCP_AUTH_TOKEN=my-secret npm run start:http
Then add the token in the integration config as a Bearer token.
Optional: Run on a remote server
If you want it always-on (e.g. on a VPS or Render/Railway):
- Deploy the project to your server
- Run
npm run start:httpwith your env vars - Use
https://your-server.com/sseas the integration URL - Set
MCP_AUTH_TOKENfor security
Keep it running (local)
Use pm2 or a simple background process:
# Install pm2 globally
npm install -g pm2
# Start as a daemon
ODDS_API_KEY=xxx SPORTMONKS_TOKEN=xxx pm2 start dist/server-http.js --name sports-mcp
# Auto-restart on reboot
pm2 startup
pm2 save
For Claude Desktop (local app)
See the config under Quick Start above ā uses stdio transport via dist/index.js.
For Claude Code
claude mcp add sports-data node /FULL/PATH/TO/sports-mcp-server/dist/index.js \
-e ODDS_API_KEY=your-key \
-e SPORTMONKS_TOKEN=your-token
Tool Reference
The Odds API Tools (prefix: odds_)
| Tool | Description | Quota Cost |
|---|---|---|
odds_get_sports |
List all available sports and their keys | Free |
odds_get_events |
List upcoming/live events for a sport | Free |
odds_get_odds |
Primary tool ā get odds for upcoming/live games (ML, spreads, totals) | markets Ć regions |
odds_get_scores |
Live scores + recently completed games | 1-2 |
odds_get_event_odds |
Odds for a single event ā supports ALL markets including player props | unique markets Ć regions |
odds_get_event_markets |
Discover available market keys for an event | 1 |
odds_get_participants |
List teams/players for a sport | 1 |
odds_get_historical_odds |
Historical odds snapshot at a timestamp | 10 Ć markets Ć regions |
odds_get_historical_events |
Historical events at a timestamp | 1 |
odds_get_historical_event_odds |
Historical odds for single event ā all markets | 10 Ć markets Ć regions |
Common sport keys: basketball_nba, basketball_ncaab, americanfootball_nfl, americanfootball_ncaaf, icehockey_nhl, soccer_epl, soccer_germany_bundesliga, soccer_spain_la_liga, soccer_france_ligue_one, tennis_atp_*, tennis_wta_*
SportMonks Tools (prefix: sm_)
| Tool | Description |
|---|---|
sm_livescores |
Live soccer scores (inplay / all today / latest updated) |
sm_fixtures |
Fixtures by ID, date, date range, team, H2H, or search |
sm_leagues |
Leagues ā all, by ID, live, by country, by team, search |
sm_seasons |
Seasons ā all, by ID, by team, search |
sm_standings |
League tables ā by season, round, or live |
sm_teams |
Teams ā all, by ID, country, season, search |
sm_players |
Players ā by ID, country, search |
sm_team_squads |
Team squad (current or by season) |
sm_statistics |
Season/stage/round statistics for teams or players |
sm_topscorers |
Top scorers by season or stage |
sm_predictions |
Probabilities and value bets (SM's own model) |
sm_expected_xg |
Expected goals (xG) by team or player |
sm_odds_prematch |
Pre-match odds by fixture, bookmaker, or market |
sm_odds_inplay |
In-play odds by fixture, bookmaker, or market |
sm_schedules |
Match schedules by season, team, or both |
sm_rounds |
Round data by season or ID |
sm_markets |
List available betting markets |
sm_bookmakers |
List available bookmakers |
Using Includes (SportMonks)
SportMonks tools support an include parameter for enriched responses:
include: "participants,scores,league,statistics,events"
include: "player,position,teams"
include: "fixtures,rounds"
Using Filters (SportMonks)
Some endpoints support filters:
filters: "fixtureLeagues:8" # Filter by league ID
filters: "eventTypes:14,18" # Filter event types
Example Prompts
Once configured, you can ask Claude things like:
- "Get me tonight's NBA odds from DraftKings and FanDuel ā spreads, moneylines, and totals"
- "Pull EPL standings for this season"
- "What are the live soccer scores right now?"
- "Compare the historical odds movement for the Lakers game last Tuesday"
- "Get Arsenal's xG data and current squad"
- "Show me value bets from SportMonks predictions for today's fixtures"
- "Get player props available for tonight's NFL game"
Architecture
src/
āāā index.ts # MCP server entry ā stdio transport (Claude Desktop / Claude Code)
āāā server-http.ts # MCP server entry ā HTTP/SSE transport (claude.ai web)
āāā odds-api.ts # The Odds API tool definitions (10 tools)
āāā sportmonks.ts # SportMonks tool definitions (18 tools)
āāā http.ts # Shared HTTP utility (fetch + URL builder)
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.