credentials

Installation
SKILL.md

Credentials

Help the user safely manage API keys and understand what data access they have.

First: check profile

Read .nutmeg.user.md. If it doesn't exist, tell the user to run /nutmeg:init first.

Key management

API keys should NEVER be committed to git. Guide the user to store them in:

  1. .env file (gitignored) for project-level keys
  2. Environment variables for system-level keys
  3. .nutmeg.credentials.local (gitignored) for nutmeg-specific config

When the user wants to add a key, help them:

  • Create/update .env with the key
  • Verify .gitignore includes .env and *.local
  • Test the key works with a minimal API call

Provider access reference

Free (no key needed)

Source Access method What you get
StatsBomb open data GitHub download / statsbombpy Event data for select competitions (World Cups, FA WSL, some league seasons)
FBref Web scraping (soccerdata) Season aggregates: passing, shooting, defense, GCA, possession stats
Understat Web scraping (soccerdata) Per-match xG, shot-level data. Top 5 European leagues
ClubElo HTTP API (no key) Historical Elo ratings for European clubs
football-data.co.uk CSV download Historical match results with betting odds
Transfermarkt Web scraping Transfer values, squad info (fragile, terms may prohibit)

Paid / keyed

Provider Free tier? Key type Env var convention
SportMonks Yes (limited) API token SPORTMONKS_API_TOKEN
Opta/Perform No (unofficial feed key exists) Feed token OPTA_FEED_TOKEN
StatsBomb API No API key + password STATSBOMB_API_KEY, STATSBOMB_API_PASSWORD
Wyscout No API key WYSCOUT_API_KEY

Data marketplaces

Source What to look for
Kaggle Search "football events", "soccer xG". Check license (many are CC-BY). Verify freshness.
GitHub Search "football data", "soccer analytics". StatsBomb open-data is the gold standard.
data.world Occasional football datasets. Check quality and documentation.

When the user asks "what can I access for free?"

Based on their .nutmeg.user.md profile, give them a concrete plan:

  • Beginners: Start with StatsBomb open data. It's the best-documented, highest-quality free source.
  • Python users: Install statsbombpy and soccerdata for immediate access to open data + FBref/Understat.
  • R users: Install StatsBombR and worldfootballR for similar coverage.
  • JS/TS users: StatsBomb open data can be fetched directly from GitHub as JSON. FBref requires scraping.

Warnings

  • Scraping websites may violate their terms of service. Advise caution.
  • Commercial API keys often prohibit data redistribution. Check the license.
  • Free tier rate limits can be very low (SportMonks free: 5 requests/day on some plans).
  • Never print or log API keys. Use process.env or equivalent.
Weekly Installs
1
GitHub Stars
17
First Seen
Mar 25, 2026