World Cup Sports Oracle

World Cup Sports Oracle

For a feature that moves money based on results, no single free source is good enough. You want a primary + fallback + sanity check, and only settle when ≥2 agree. This is why we designed a custom 3-Source Consensus Sports Oracle.

Architecture

MoniBot queries three independent data layers:

1. Primary Source: football-data.org
2. Fallback Source: API-Football (api-sports.io)
3. Sanity Check: openfootball (on GitHub)

Ten minutes after a match's official completion, the oracle evaluates the results across all three sources.

graph TD
    A[Football-Data.org API] --> D{Consensus Engine}
    B[API-Football API] --> D
    C[OpenFootball GitHub] --> D
    D -->| >= 2 agree | E[Execute/Cancel Payment]
    D -->| Disagreement | F[Dispute Safety Lock]

Consensus Engine

To prevent erroneous payouts from a single corrupted data feed:

• Majority Agreement: At least 2 out of the 3 data sources must agree on the final match status and scores.
• Name Normalisation: Resolves spelling variations and accents across sources (e.g. Curaçao -> curacao) to ensure match pairings align correctly.
• Dispute Safety Lock: If the sources disagree, the transaction is marked as "disputed" in the database, halting all automated execution and triggering a notification for administrative review.