jk-mcp-usls
# jk-mcp-usls
MCP server that gives Claude live access to USL Super League data — teams, matches, standings, rosters, and schedule-strength analytics — via the ESPN public API.
[](https://github.com/jedi-knights/jk-mcp-usls/actions/workflows/ci.yml)
[](https://github.com/jedi-knights/jk-mcp-usls/actions/workflows/badge.yml)
[](https://jedi-knights.github.io/jk-mcp-usls/)
[](https://github.com/jedi-knights/jk-mcp-usls/actions/workflows/evals.yml)
[](https://github.com/jedi-knights/jk-mcp-usls/actions/workflows/release.yml)
[](https://www.python.org/downloads/)
[](LICENSE)
---
## Table of Contents
- [Overview](#overview)
- [Features](#features)
- [Requirements](#requirements)
- [Installation](#installation)
- [Usage](#usage)
- [Configuration](#configuration)
- [Claude Code](#claude-code)
- [Claude Desktop](#claude-desktop)
- [Docker](#docker)
- [Development](#development)
- [Contributing](#contributing)
- [License](#license)
---
## Overview
AI assistants like Claude are knowledgeable, but they have a hard cutoff date — they cannot tell you today's USL Super League standings, last night's scores, or which teams are currently in a playoff position. This project fixes that.
It is an **MCP server** — a plugin that gives Claude direct access to live USL Super League data: scores, standings, rosters, and derived schedule-strength analytics. Once installed, you can ask Claude natural-language questions about the USL Super League and get accurate, up-to-date answers. No subscription, no API key, and no programming required to use it.
This is the **v1 scaffold** — it wraps the ESPN public API only. The league's own site (`gainbridgesuperleague.com`) exposes match data only through a licensed Opta widget embed, so ESPN is the only clean JSON source. Cup competitions and richer stats are on the roadmap if a stable second-tier feed becomes available.
---
## Features
The v1 surface is eleven read-only, idempotent tools split across two tiers.
### ESPN-backed (8)
| Tool | Description |
|---|---|
| `get_teams` | List all 8 USL Super League clubs with IDs and abbreviations |
| `get_team` | Details for a specific team |
| `get_roster` | Team's active roster — jersey, position, age, citizenship |
| `get_scoreboard` | Match scores for a single day, a date range, or the current matchweek |
| `get_team_schedule` | Every match for a team in the current season — past + upcoming |
| `get_match_details` | One match's full details — score, venue, attendance, goals, cards, subs |
| `get_standings` | Current standings — single 8-team table ordered by points |
| `get_news` | Recent USL Super League news articles |
### Derived analytics (3)
Pure functions over live standings + team schedules, exposing schedule-strength context the raw table does not.
| Tool | Description |
|---|---|
| `get_strength_of_schedule` | Team's average opponent points-per-game across matches already played |
| `get_results_by_opponent_tier` | Team's W-L-T split across current top / middle / bottom standings tiers |
| `get_adjusted_points_per_game` | Team's raw PPG alongside an opponent-quality-adjusted PPG |
### Roadmap
Deferred to v2+:
- Player leaderboards and team season aggregates if a stable USL SL Opta feed becomes accessible (today the league's Opta widget is subscription-gated)
- Press-release feed from `gainbridgesuperleague.com/wp-json/wp/v2/sec_news`
- Playoff bracket rendering
- Related women's competitions the league may add (Concacaf W Champions Cup, USL Cup)
---
## Requirements
- [Python 3.13+](https://www.python.org/downloads/)
- [uv](https://docs.astral.sh/uv/getting-started/installation/)
---
## Installation
```bash
git clone https://github.com/jedi-knights/jk-mcp-usls.git
cd jk-mcp-usls
uv sync
```
---
## Usage
Run the server in stdio mode (the default — used by Claude Code and Claude Desktop):
```bash
uv run python -m usls.server
```
Run in HTTP mode (for networked or deployed access):
```bash
MCP_TRANSPORT=streamable-http uv run python -m usls.server
```
### Example prompts
**Standings, scores, rosters:**
- Who is leading the USL Super League right now?
- Show me every USL Super League result from this past weekend.
- Who is on Brooklyn FC's roster?
- When does Carolina Ascent play next?
**Schedule strength:**
- Which USL Super League team has played the toughest schedule so far?
- Show me Brooklyn FC's record against the current top 3 teams.
- Compare Carolina Ascent and DC Power on adjusted points-per-game.
---
## Configuration
All configuration is via environment variables. None are required for local use.
| Variable | Default | Description |
|---|---|---|
| `MCP_TRANSPORT` | `stdio` | Transport mode: `stdio` or `streamable-http` |
| `HOST` | `0.0.0.0` | Bind address (HTTP transport only) |
| `PORT` | `8000` | TCP port (HTTP transport only) |
| `MCP_PATH` | `/mcp/usls` | URL path (HTTP transport only) |
| `API_HOST` | `https://site.api.espn.com` | ESPN API base URL |
| `LOG_LEVEL` | `INFO` | `DEBUG`, `INFO`, `WARNING`, or `ERROR` |
| `MCP_TRACING_ENABLED` | unset | Bootstrap the OpenTelemetry SDK |
| `MCP_AUTH_ENABLED` | unset | Require RS256 bearer tokens on streamable-http |
| `MCP_AUTH_ISSUER_URL` | unset | Auth-server origin (required when auth is on) |
| `MCP_AUTH_RESOURCE_URL` | unset | This server's public URL for the `aud` claim |
---
## Claude Code
Install from your local clone globally so the server is available in every project:
```bash
claude mcp add --scope user usls -- uv run --directory /path/to/jk-mcp-usls python -m usls.server
```
Replace `/path/to/jk-mcp-usls` with the absolute path to your clone. Verify with `claude mcp list`.
Drop `--scope user` to register only for the current project, or commit a `.mcp.json` to the repo root for collaborators:
```json
{
"mcpServers": {
"usls": {
"command": "uv",
"args": ["run", "--directory", "/path/to/jk-mcp-usls", "python", "-m", "usls.server"]
}
}
}
```
---
## Claude Desktop
Add the following to your Claude Desktop configuration file.
**Location:**
- macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`
- Windows: `%APPDATA%\Claude\claude_desktop_config.json`
```json
{
"mcpServers": {
"usls": {
"command": "uv",
"args": [
"run",
"--directory", "/path/to/jk-mcp-usls",
"python", "-m", "usls.server"
]
}
}
}
```
If `uv` is not on Claude Desktop's `PATH`, use the absolute path (`which uv` will show it). Fully quit and relaunch Claude Desktop after saving — a window close is not enough.
---
## Docker
Build the image:
```bash
docker build -t jk-mcp-usls:latest .
```
Run in stdio mode (for MCP clients that spawn a subprocess):
```bash
docker run -i --rm jk-mcp-usls:latest
```
Run in HTTP mode:
```bash
docker run --rm -p 8000:8000 \
-e MCP_TRANSPORT=streamable-http \
jk-mcp-usls:latest
```
---
## Development
### Install
```bash
uv sync
```
### Invoke tasks
All common workflows are `invoke` tasks. Run `uv run inv --list` to see everything.
| Task | Alias | Description |
|---|---|---|
| `uv run inv lint` | `inv l` | Run ruff linter and format check |
| `uv run inv lint --fix` | `inv l --fix` | Auto-fix lint violations and reformat |
| `uv run inv test` | `inv t` | Run the full test suite |
| `uv run inv coverage` | `inv v` | Run tests with coverage report (threshold: 90%) |
| `uv run inv check-complexity` | `inv cc` | Check cyclomatic complexity (max 7) |
| `uv run inv build` | `inv b` | Build wheel and sdist into `dist/` |
| `uv run inv build-image` | `inv bi` | Build the Docker image |
| `uv run inv clean` | `inv c` | Remove build and coverage artifacts |
### Project structure
```
src/usls/
├── server.py # entry point, transport selection, logging setup
├── adapters/
│ ├── inbound/
│ │ ├── mcp_adapter.py # FastMCP server, health endpoints, tool registration
│ │ ├── formatters.py # domain → LLM-readable text
│ │ ├── authorization.py # inbound authz port implementations
│ │ └── tools/
│ │ ├── espn.py # 8 ESPN-backed tools
│ │ └── analytics.py # 3 schedule-strength analytics tools
│ └── outbound/
│ ├── espn_adapter.py # ESPN HTTP client
│ ├── parsers.py # ESPN JSON → domain models
│ ├── retry_adapter.py # transient-failure retry decorator
│ └── caching_adapter.py # in-process TTL cache
├── application/
│ ├── service.py # USLSService — use cases, orchestration
│ ├── _helpers.py # input validation
│ └── _analytics_helpers.py # pure math for schedule-strength tools
├── domain/
│ ├── models.py # Team, Match, Standing, etc.
│ └── exceptions.py # USLSNotFoundError, UpstreamAPIError
├── ports/
│ ├── inbound.py # Authorizer protocol
│ └── outbound.py # USLSAPIPort protocol
├── observability/ # OpenTelemetry bootstrap (opt-in)
└── security/ # JWKS token verifier
```
The dependency direction flows inward: adapters → ports → domain. Nothing in `domain/` imports from adapters or a framework.
---
## Contributing
1. Fork the repository and clone your fork
2. Create a feature branch: `git checkout -b feature/your-feature`
3. Make your changes following the existing patterns (hexagonal architecture, TDD, conventional commits)
4. Verify the full check suite passes: `uv run inv lint && uv run inv check-complexity && uv run inv coverage`
5. Open a pull request against `main`
All CI checks (lint, complexity, tests, coverage ≥ 90%) must pass before merge.
---
## License
MIT — see [LICENSE](LICENSE).
TDQS
Scored across 11 tools
Each tool targets a distinct data type or query: team lists vs. details, schedules vs. scoreboards vs. match events, and three separate analytics tools that do not overlap. An agent can easily select the right tool without confusion.
Every tool follows the same get_ prefix with descriptive noun phrases (get_teams, get_standings, get_match_details). This consistent pattern makes the toolset predictable and easy to navigate.
With 11 tools, the server is well-scoped for a sports data API: core retrieval (teams, standings, schedules, scores, rosters, news) plus a few advanced analytics tools. No tool feels redundant or out of place.
The domain of USL Super League data is thoroughly covered: teams, standings, matches, rosters, news, and analytical queries. The read-only nature is consistent with a data feed, and no critical data type appears missing.