Data Lineage Access Governance

This simulator renders a live data-lineage graph — source datasets feeding derived datasets feeding reports — and runs two governance mechanisms on it in real time. Sensitivity classification propagates downstream: a derived dataset must inherit at least the strictest classification of everything it was built from, and any node that doesn't is flagged as a live audit violation. Around every dataset, small markers represent active access grants that age over simulated time; a periodic recertification review sweeps them on a configurable interval, renewing each with a tunable probability and letting the rest drift to stale, then revoked, exactly like a real access-certification campaign. Tune the review interval, recertification rate and simulated time speed, grant new access, extend the lineage with derived datasets, and watch the privilege-creep index track how much entitlement drift accumulates between audits.