Architecture overview

Evidence → Meaning → Stability → Authorization → Execution → Proof. SPP2 is responsible for semantic evidence and assertion support only. Governance layers (stability/authorization) are independently evaluated by separate engines whose verdicts are shown alongside SPP2's, never merged into it.

Your application
   → SPP2 Gateway (auth, quota, metering)
      → SPP2 Core (semantic evaluation, unmodified validated engine)
         → optional governance probes (independently attributed, never merged)
      → Receipt Vault (durable, tamper-evident)
   → your webhook / dashboard

Proprietary engine internals (scoring logic, thresholds, calibration) are not documented here — see Security model for what is and isn't exposed.