Docs

Oracle dependency

MAOB is anchored to an oracle midpoint via IOracleAdapter.getPrice(address orderBook).

Risks:

• Staleness: If oracle updates stop, taker/maker actions that require price can revert (staleness is enforced inside the OracleAdapter per leg).
• Out-of-range: MAOB performs overflow safety checks and can revert with MAOB__OracleOutOfRange() for extreme values.
• Incorrect pricing: If the oracle midpoint is manipulated or incorrect, rung pricing and fills will reflect that.

Mitigations:

• Use a robust oracle and configure appropriate per-leg staleness thresholds in the OracleAdapter.

Pausing controls

MAOB uses a market state machine:

• setMarketState(MarketState, reason) sets one of Active, Halted (makers + takers paused), or UnwindOnly (makers paused, takers active).
• UnwindOnly is time-bound and auto-expires back to effective Active.

Operationally, this provides an explicit emergency mode while preventing unsafe asymmetric states.

Rounding and dust

MAOB uses:

• token scaling (baseScale, quoteScale)
• significant-figure rounding for internal price values
• remainder-carrying arithmetic for fair distribution across many fills

Queue and fill accounting assumptions

MAOB’s order accounting uses per-rung “generations” and a Fenwick-tree based queue model. This is designed to:

• preserve fairness for makers within a rung
• allow efficient partial fills and claims

Assumptions:

• Makers are responsible for claiming; output does not automatically stream to wallets.

CLP operational risk

CLP is a vault that depends on an operator. Risks include:

• Strategy risk: poor rung placement can lead to adverse selection or inventory imbalance.
• Operational delays: if the operator does not regularly claim/cancel/redeploy, the vault may hold more capital in less-liquid states.

Mitigations:

• Establish operator procedures and monitoring.
• Use conservative withdrawal cooldown/fees if needed.