Newton Protocol is a decentralized policy protocol for runtime invariant enforcement. It provides a security layer for on-chain systems that cannot afford silent failure—stopping invariant violations before execution rather than detecting them after damage occurs.
Static audits verify intent, but attackers exploit edge-case execution. Most DeFi exploits happen not because teams forgot a check, but because assumptions silently failed at runtime. Newton addresses this gap:
- Preventative, not reactive: Transactions that violate invariants never execute. No state changes, no funds move.
- Context-aware verification: Policies can incorporate off-chain signals (oracle prices, market volatility, risk scores) evaluated via secure multi-party computation.
- No contract rewrites: Integration requires a single policy-verification hook on sensitive operations. Your core logic remains unchanged.
Numerical & Economic Invariants Pool value bounds, precision loss thresholds, collateral ratios, liquidity requirements, oracle deviation tolerance.
Market-Condition Invariants Automatic circuit breakers based on volatility, price deviation, or stress scenarios—without pausing the protocol.
Behavioral & Authorization Invariants Compromised key detection, insider drain prevention, sanctioned flow blocking, abnormal transaction pattern rejection.
Policies are evaluated by a decentralized, permissioned operator set. No single operator can approve execution, eliminating the centralized security admin key problem. Every successful evaluation produces a cryptographic proof that the policy was satisfied at a specific time for a specific operation, providing deterministic enforcement and auditable guarantees.
For protocols that rely on precision math, liquidity assumptions, oracles, complex state transitions, or cross-contract composition—Newton turns invariants from audit notes into enforced rules.
To learn more, read the Litepaper or check out our docs.
If you've found a security vulnerability in a Newton project, please see SECURITY.md.