MCS-REB-1

What is checked, and to what tolerance

Every bound carries a reason, not just a number. A tolerance nobody can justify is a tolerance that will be widened the first time it is inconvenient.

• currentTickThe tick is a single integer read from slot0 at a fixed block. There is no ambiguity and therefore no tolerance.
• proposedTickSpacingThe pool contract reverts on a tick that is not a multiple of its spacing, so a plan that violates this is not merely suboptimal, it is unexecutable. This is the single highest-signal check in the set and the most common failure in naive agents.
• policyRangeThere is no objectively correct V3 range in the abstract, so the case supplies the policy and we check compliance with it. One spacing of slack is the minimum the pool’s own granularity permits: the ideal tick is rarely itself a valid multiple.
• amountsThe formula is exact, but agents legitimately round for slippage and gas. 0.5% admits sane rounding while rejecting an agent that has not actually run the maths.
• inRangeDerived from tickLower <= tick < tickUpper. One right answer.
• distanceToBoundDerived arithmetic from the tick and the two bounds. The half-point band absorbs differences in rounding convention, not differences in method.
• slippageBoundAn execution plan with no slippage bound is unsafe to sign regardless of how good its range is. We check only that a bound is stated, never whether the chosen value was wise — that is a Ledger question.

The ground-truth method

The expected values are computed by us from chain state at a pinned block, never taken from a third party. A case is captured immediately before a run because BSC keeps roughly 64 blocks of state — about 29 seconds — so a case captured earlier could not be read by the agent under test or by us.

• REB-1-btcb-usdc-2500subject {"owner":"0x2e07E0145C0CFdF6D200B0aFAeD36953ef00d0cD","tokenId":"7321916","pair":"BTCB/USDC"} · policy {"rangePct":6,"feeTier":2500,"statement":"re-centre the position symmetrically at ±6% around the current spot price, on the 0.25% fee tier, keeping the same liquidity"}

What this test refuses to grade

• Whether the chosen range is the wisest availableThere is no objectively correct V3 range in the abstract. The test case supplies a policy and we check compliance with it; the quality of the policy is a Ledger question.

A standard is defined as much by what it refuses to grade as by what it grades. Every one of these is a judgement question and belongs in the Ledger, against a rubric registered before anyone has seen an answer.

Results, including the failures

TESTED1 pass, 99 fail, across 100 recorded runs. Failures are listed first because they are the more useful half.

Failing to answer and answering wrongly are different facts and are shown as such. Test stubs are excluded from this list.