CAIN-42 CAIN Studio
25CAIN-42 Top 25 › ACTUARY › Attack lab

Attack lab

Score real fleets, then thin out the evidence and see whether the score notices.

5 attacks 2 controls live: nothing pre-recorded

  1. Pick a preset. Red ones are attacks, green ones are honest requests that must still work.
  2. Tamper. Choose any field and drop it, null it, change its type, upper-case it, flip one character, or add a field the product never asked for.
  3. Send and read the real reply, or run every preset and watch the scoreboard fill.

What each preset tests

PresetRouteWhy it matters
attackSample fleet: Accounts-payable agent: wire transfers and bank-detail changes ran ungatedPOST /actuary-keystone/scanA fail-closed risk score computed from CAIN decision evidence.
attackSample fleet: Coding agent: shell commands and a branch deletion ran without a humanPOST /actuary-keystone/scanA fail-closed risk score computed from CAIN decision evidence.
controlSample fleet: Support agent: every consequential action held or blocked by CAINPOST /actuary-keystone/scanA fail-closed risk score computed from CAIN decision evidence.
controlSample evidence, unchangedGET /actuary-keystone/samples/governed-support-fleet/evidence → POST /actuary-keystone/scan-evidenceThe same evidence through the free-form route scores the same.
attackSame evidence with every decision removedGET /actuary-keystone/samples/governed-support-fleet/evidence → POST /actuary-keystone/scan-evidenceZero decisions is NO_EVIDENCE, not a good score.
attackOne decision only (thin evidence)GET /actuary-keystone/samples/governed-support-fleet/evidence → POST /actuary-keystone/scan-evidenceToo little evidence cannot earn LOW_RISK.
attackEvidence with no decisions listPOST /actuary-keystone/scan-evidenceMalformed evidence is refused, not scored.

Tamper with any field and send it again: a verifier must answer no, never crash or say yes. Public demo routes are rate limited per address; a 429 shows as rate limited. Our own lab; no independent test.

FIRME AI CORPORATION · CAIN-42 is a Byzantine trust fabric. Numbers marked "our own check" were run by us, not by an independent party. Nothing on these pages is a certification.