MISBEHAVEINCENTIVE QA

Find the loophole in your incentive — before your customers do.

Challenger agents stress-test your rule, find the shortest path to payout without proof of value, and rerun the same tests against a repair.

GAP
PAYOUTwhere your rule paysVALUEwhere your business wins

Generates adversarial tests, not predictions of behavior.

Do not enter confidential, personal, or production customer information. Live runs are stored so their replay link keeps working; curated examples stay in your browser.

HOW IT WORKS

One question, asked four ways.

Can the payout be reached without proving value? Everything below is machinery for answering that.

  1. 01

    You write the rule.

    Plain language, exactly as it would ship — plus the one event that proves the business got what it paid for.

  2. 02

    Four challengers try to break it.

    A1 EXPECTED completes the intended journey. A2 SHORTEST PATH looks for the fewest actions that still collect. A3 REPEAT tries to do it again. A4 CONTROL checks whether the written rule grants anyone the authority to stop it. They run one at a time, against your text.

  3. 03

    You get a counterexample — or you get told there isn't one.

    A run that receives the reward without ever proving value stops the bench and shows the shortest path to it. If no challenger finds one, the verdict says so. The product never manufactures a finding.

WHAT YOU GET

A decision memo, and the tests behind it.

Every run ends the same way: the finding, what the written text mechanically permits, a repair with exactly three guardrails and their costs, and the same four tests rerun against the repaired rule.

THE REFERRAL EXAMPLE — ACTUAL OUTPUT
MISBEHAVEINCENTIVE QA

CONTROLLED PILOT RECOMMENDED

Passing under the current test suite.

The repeatable payout path no longer completes.

The expected journey remains eligible.

Introduced friction

  • Identity verification
  • 30-day payout delay
  • Household exception review

RESIDUAL RISK

Household members sharing an address still require a human review decision.

REGRESSION TEST SUITE

The same four tests, rerun against the repaired rule.

T-01Expected journeyPASS
T-02Payout without proof of valueBLOCKED
T-03Repeatable payout pathBLOCKED
T-04Household edge caseREVIEW

WHY TRUST IT

The honesty rules are the product.

An incentive review that overstates itself is worse than none, because it gets believed. These are load-bearing, not marketing.

Adversarial tests, not predictions.
MISBEHAVE generates plausible ways a rule can be satisfied without value being created. It does not forecast what real people will do, and it never claims to.
Permitted exposure, not predicted loss.
The exposure figure is arithmetic on the rule's own number, scaled over 1, 10 or 100 events. It says what the written text mechanically permits — nothing about likelihood.
No invented numbers.
No probabilities, no confidence percentages, no exploitability score, no gauges. Only counts that trace back to a generated test.
Never “safe”.
The best verdict this product will give you is “passing under the current test suite” — because four tests cannot establish anything stronger.
Every quotation is verified.
A cited phrase is checked as an exact substring of the rule you submitted before it is ever displayed.
Repairs show their cost.
Each guardrail carries its tradeoff. A patch is never presented as free.