@cyanheads/sports-mcp-server
Get live scores, schedules, standings, team and player data for NFL, NBA, MLB, NHL, soccer, and more via MCP.
README
<div align="center"> <h1>@cyanheads/sports-mcp-server</h1> <p><b>Get live scores, schedules, standings, team and player data for NFL, NBA, MLB, NHL, soccer, and more via MCP. STDIO or Streamable HTTP.</b> <div>7 Tools</div> </p> </div>
<div align="center">
</div>
<div align="center">
</div>
Tools
Seven tools organized around what an agent needs — find a team or player, get today's scores, check the schedule, read the standings, or pull full team or player detail:
| Tool | Description |
|---|---|
sports_find_team |
Resolve a team name or partial name to its canonical record and source IDs. Use before any team-scoped query. |
sports_find_player |
Resolve a player name to their canonical record via TheSportsDB. Disambiguation step before player-scoped queries. |
sports_get_scores |
Live and final scores for a league on a given date, optionally scoped to a specific team. |
sports_get_schedule |
Upcoming and past fixtures for a team or league over a date range. |
sports_get_standings |
Current standings or league table for a league and season. |
sports_get_team |
Team detail: active roster, last 5 results, next 3 fixtures, venue, and metadata. |
sports_get_player |
Player detail: bio, current team, position, nationality, birth date, height/weight, and thumbnail. |
sports_find_team
Resolve a fuzzy team name to its canonical record across ESPN, MLB StatsAPI, and TheSportsDB.
- Returns full name, league, logo URL, venue, and ESPN/MLB/TheSportsDB cross-reference IDs
- Use before
sports_get_scores,sports_get_schedule,sports_get_standings, orsports_get_teamto get a validteam_name - Fuzzy match on display name, abbreviation, or location (e.g. "Mariners", "SEA", "Seattle Seahawks")
sports_find_player
Resolve a player name to their canonical record.
- Multi-sport player search via TheSportsDB
- Optional
sportfilter to narrow ambiguous names (e.g. "Michael Jordan") - Returns player ID, full name, current team, position, nationality, birth date, and thumbnail URL
- Use the returned
player_idwithsports_get_playerfor full bio detail
sports_get_scores
Live and final scores for a league on a given date.
- Routes NFL/NBA/NHL/soccer → ESPN; MLB → MLB StatsAPI (more authoritative)
- Returns home/away teams, current score, status (
scheduled/in-progress/final), period/clock, and UTC start time - Omit
datefor today's games; useteam_nameto filter to one team's game - Returns
games: [], reason: '...'(not an error) when no games are scheduled
sports_get_schedule
Upcoming and past fixtures for a team or league over a date range.
- Fetches full season from ESPN and applies
date_from/date_tofiltering server-side - Returns opponent, home/away flag, UTC date/time, venue, and result for completed games
- Omit
team_namefor the full league calendar; provide it for a single team's fixtures
sports_get_standings
Current standings or league table for a league and season.
- Returns rank, W/L (or points for soccer/NHL), division/conference, streak, and games behind
- Omit
seasonfor the current season; pass a YYYY year for historical standings - NHL uses points system (
wins/otLosses/losses); soccer returnspointsfor the league table
sports_get_team
Composite team detail combining multiple source calls.
- Active roster (or squad), last 5 results, next 3 fixtures, venue, and team metadata
- MLB teams use MLB StatsAPI for roster and schedule; all others use ESPN
sports_get_player
Full player profile from TheSportsDB.
- Bio, current team, position, nationality, birth date, height/weight, career description
- Media thumbnail URL for display
- Accepts
tsdb:-prefixed IDs (fromsports_find_player) or raw numeric TheSportsDB IDs
Features
Built on @cyanheads/mcp-ts-core:
- Declarative tool definitions — single file per tool, framework handles registration and validation
- Unified error handling — handlers throw, framework catches, classifies, and formats
- Pluggable auth:
none,jwt,oauth - Swappable storage backends:
in-memory,filesystem,Supabase,Cloudflare KV/R2/D1 - Structured logging with optional OpenTelemetry tracing
- STDIO and Streamable HTTP transports
Sports-specific:
- Three keyless sources — ESPN site API, MLB StatsAPI, TheSportsDB (free tier key
3) — no API credentials required - League routing table internally routes each query to the best source per sport; agents never reference an upstream API
- Normalized output types across all sources —
NormalizedGame,NormalizedTeam,NormalizedPlayer,NormalizedStanding— with source provenance on every record - Graceful degradation — empty scoreboards (off-season, no games) return
games: []with areasonstring, not an error
Agent-friendly output:
- Source provenance on every record —
source: 'espn' | 'mlbstats' | 'thesportsdb'so agents can reason about data authority - Structured
reasonfield on empty score responses so agents can explain the result to users - Cross-source IDs surfaced by
sports_find_team—espnId,mlbId,tsdbId— for seamless downstream routing without re-resolving names
Getting started
Add the following to your MCP client configuration file.
{
"mcpServers": {
"sports": {
"type": "stdio",
"command": "bunx",
"args": ["@cyanheads/sports-mcp-server@latest"],
"env": {
"MCP_TRANSPORT_TYPE": "stdio",
"MCP_LOG_LEVEL": "info"
}
}
}
}
Or with npx (no Bun required):
{
"mcpServers": {
"sports": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@cyanheads/sports-mcp-server@latest"],
"env": {
"MCP_TRANSPORT_TYPE": "stdio",
"MCP_LOG_LEVEL": "info"
}
}
}
}
Or with Docker:
{
"mcpServers": {
"sports": {
"type": "stdio",
"command": "docker",
"args": [
"run", "-i", "--rm",
"-e", "MCP_TRANSPORT_TYPE=stdio",
"ghcr.io/cyanheads/sports-mcp-server:latest"
]
}
}
}
For Streamable HTTP, set the transport and start the server:
MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
# Server listens at http://localhost:3010/mcp
Prerequisites
- Bun v1.3.0 or higher (or Node.js v24+).
- No API keys required — ESPN and MLB StatsAPI are fully keyless; TheSportsDB ships with a free public test key (
3).
Installation
- Clone the repository:
git clone https://github.com/cyanheads/sports-mcp-server.git
- Navigate into the directory:
cd sports-mcp-server
- Install dependencies:
bun install
- Configure environment:
cp .env.example .env
# edit .env to override defaults (all optional)
Configuration
| Variable | Description | Default |
|---|---|---|
THESPORTSDB_API_KEY |
TheSportsDB API key. Replace with a paid key for higher rate limits. | 3 |
MCP_TRANSPORT_TYPE |
Transport: stdio or http. |
stdio |
MCP_HTTP_PORT |
Port for HTTP server. | 3010 |
MCP_AUTH_MODE |
Auth mode: none, jwt, or oauth. |
none |
MCP_LOG_LEVEL |
Log level (RFC 5424). | info |
LOGS_DIR |
Directory for log files (Node.js only). | <project-root>/logs |
OTEL_ENABLED |
Enable OpenTelemetry instrumentation. | false |
See .env.example for the full list of optional overrides.
Running the server
Local development
-
Build and run:
# One-time build bun run rebuild # Run the built server bun run start:stdio # or bun run start:http -
Run checks and tests:
bun run devcheck # Lint, format, typecheck, security bun run test # Vitest test suite bun run lint:mcp # Validate MCP definitions against spec
Docker
docker build -t sports-mcp-server .
docker run --rm -p 3010:3010 sports-mcp-server
The Dockerfile defaults to HTTP transport, stateless session mode, and logs to /var/log/sports-mcp-server. OpenTelemetry peer dependencies are installed by default — build with --build-arg OTEL_ENABLED=false to omit them.
Project structure
| Directory | Purpose |
|---|---|
src/index.ts |
createApp() entry point — registers tools and inits services. |
src/config |
Server-specific environment variable parsing (THESPORTSDB_API_KEY). |
src/services/types.ts |
Normalized cross-source types and league routing table. |
src/services/espn |
ESPN site API service — scores, schedules, standings, teams. |
src/services/mlb |
MLB StatsAPI service — scores, schedules, standings, rosters. |
src/services/thesportsdb |
TheSportsDB service — player and team search/metadata. |
src/mcp-server/tools |
Tool definitions (*.tool.ts). |
tests/ |
Unit and integration tests mirroring src/. |
docs/ |
Design doc and directory tree. |
Development guide
See CLAUDE.md/AGENTS.md for development guidelines and architectural rules. The short version:
- Handlers throw, framework catches — no
try/catchin tool logic - Use
ctx.logfor request-scoped logging,ctx.statefor tenant-scoped storage - Register new tools via the barrel in
src/mcp-server/tools/definitions/index.ts - Wrap external API calls: validate raw → normalize to domain type → return output schema; never fabricate missing fields
Contributing
Issues and pull requests are welcome. Run checks and tests before submitting:
bun run devcheck
bun run test
License
Apache-2.0 — see LICENSE for details.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。