Pulls live FIFA World Cup 2026 data and wraps a points-based prediction game into a single script skill. You get match schedules, live scores, and results, plus functions to place bets on outcomes (win/draw/loss, exact score, total goals, first goal), cancel them before they settle, and check leaderboards sorted by points or accuracy. It's all Python imports from core.skill_tools.worldcup that return JSON, no external API keys needed. The prediction types are nicely structured, stakes are multiples of 10, and there's a full betting status check so you can confirm balance before placing. Good for World Cup chat where users want both info and a lightweight game layer without leaving the conversation.
npx -y skills add starchild-ai-agent/official-skills --skill worldcup --agent claude-codeInstalls into .claude/skills of the current project.
FIFA World Cup 2026 data + the points-based prediction game. Script skill —
call the functions in core.skill_tools.worldcup from bash and read the JSON.
Auth is automatic (container JWT); no API key needed.
python3 -c "from core.skill_tools import worldcup; import json; print(json.dumps(worldcup.get_today_matches()))"
Every function returns a dict like {"success": true, "data": ...} (or
{"success": false, "error": ...} on failure).
from core.skill_tools import worldcup)| Function | Returns |
|---|---|
get_today_matches() | today's matches + your prediction status |
get_upcoming_matches(limit=10) | next scheduled matches |
get_live_matches() | matches in progress now |
get_finished_matches(limit=20) | recent results |
get_match_detail(match_id) | one match in full |
place_bet(match_id, predict_type, prediction, stake) | places a prediction |
cancel_prediction(prediction_id) | cancels an unsettled prediction, refunds stake |
get_betting_status() | point balance, pending stakes, min/max stake |
get_my_predictions(match_id=None, settled_only=False) | your prediction history |
get_prediction_detail(prediction_id) | one prediction in full |
get_my_stats() | your win rate, total staked, etc. |
get_leaderboard(limit=50, sort_by="points") | top predictors (sort_by: points|accuracy) |
predict_type and predictionprediction is a dict whose shape depends on predict_type. Fill ONLY the keys
for that type:
| predict_type | prediction | meaning |
|---|---|---|
win_draw_loss | {"choice": "home"|"draw"|"away"} | match result |
exact_score | {"home": N, "away": N} | exact final score |
total_goals | {"range": "0-1"|"2-3"|"4+"} | total goals bucket |
first_goal | {"team": "TEAM_CODE"} | which team scores first (e.g. BRA) |
stake = points to wager: multiple of 10, min 10, max 10000.
get_betting_status() — confirm enough available points.get_match_detail(match_id) or get_today_matches() — confirm the match is open and get team codes (needed for first_goal).place_bet(match_id, predict_type, prediction, stake).Example:
python3 -c "from core.skill_tools import worldcup; import json; print(json.dumps(worldcup.place_bet(42, 'win_draw_loss', {'choice':'home'}, 100)))"
cancel_prediction only works while a prediction is unsettled; it refunds the stake.sickn33/antigravity-awesome-skills
moizibnyousaf/ai-agent-skills
github/awesome-copilot