Ecuabet

Overview

This skill uses the local scripts in scripts/ as the operational core and local docs in references/ for advanced execution patterns.

Activation Triggers

• User mentions $ecuabet.
• User gives an Ecuabet match id/url and asks for predictions.
• User asks for live refresh with cards/fouls/offsides/corners/weather/form context.
• User asks for lower-risk vs higher-return tradeoff using current market state.

Layout

• scripts/main.py: integrated entrypoint (all feeds + recommendations).
• scripts/ecuabet.py: Ecuabet markets/tracker.
• scripts/sofascore.py: live match incidents/stats.
• scripts/espn.py: ESPN summary/team stats/key events.
• scripts/open_meteo.py: weather context.
• scripts/understat.py: form/xG/season modeling.
• scripts/recommendations.py: scoring/ranking engine.
• scripts/tests/: regression tests.
• references/recipes.md: command cookbook.
• references/shortcuts.md: trigger-word prompt shortcuts.
• references/workflows.md: advanced scenario workflows.
• references/output-contract.md: JSON field map and extraction priorities.
• references/troubleshooting.md: feed failure diagnosis + recovery.

Runbook

Assumption: current directory is this skill folder (the one containing SKILL.md).

One-shot integrated output

uv run scripts/main.py <match_id_or_url> \
  --ecuabet <match_id_or_url> \
  --require-ecuabet \
  --no-raw \
  --compact

Live watch mode

uv run scripts/main.py <match_id_or_url> \
  --ecuabet <match_id_or_url> \
  --require-ecuabet \
  --watch 15 \
  --max-iterations 0 \
  --no-raw \
  --compact

Recommendation tuning

uv run scripts/main.py <match_id_or_url> \
  --ecuabet <match_id_or_url> \
  --require-ecuabet \
  --recommend-top 10 \
  --recommend-min-odds 1.01 \
  --recommend-max-odds 4.0 \
  --recommend-min-confidence 0.55 \
  --stale-threshold-seconds 120 \
  --line-history-limit 240 \
  --no-raw \
  --compact

Per-feed debug

uv run scripts/ecuabet.py <match_id_or_url> --no-raw --compact
uv run scripts/sofascore.py "<team_a> <team_b>" --no-raw --compact
uv run scripts/espn.py "<team_a> <team_b>" --league esp.1 --no-raw --compact
uv run scripts/open_meteo.py "<lat,lon>" --hourly-limit 12 --compact
uv run scripts/understat.py --league La_Liga --season 2025 --home-team "<team_a>" --away-team "<team_b>" --compact

Quality gate

cd scripts
uv run pytest tests -q

Output Focus

Primary fields to report from scripts/main.py output:

• oneShot.topRecommendation
• oneShot.shortlist
• oneShot.globalConfidence
• oneShot.feedHealth
• decisionSummary.liveMetrics
• decisionSummary.timeline
• decisionSummary.ecuabetMarkets
• feedErrors

Reference Loading Rules

• Start with references/recipes.md for exact runnable commands.
• Load references/shortcuts.md when user uses compact prompts like <id> minmax.
• Load references/workflows.md when the user asks for pre-match, in-play, or watch-loop strategy.
• Load references/output-contract.md when you need strict output-field interpretation.
• Load references/troubleshooting.md when any feed fails or data conflicts.

Resource Policy

• Use scripts/ for execution.
• Use references/ for advanced guidance and structured interpretation.
• No assets/ are required for this skill at this time.