CAIN-42 CLAWXAttack lab
The live demo compares two answer pairs from a refund-policy chatbot. Each tool checks one fact of the same live run.
2 attacks 1 controls live: nothing pre-recorded
- Pick a preset. Red ones are attacks, green ones are honest requests that must still work.
- 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.
- Send and read the real reply, or run every preset and watch the scoreboard fill.
What each preset tests
| Preset | Route | Why it matters | |
|---|---|---|---|
| attack | Overall: the model swap caused a regression | POST /catalog-demo/caindrift/run | One answer flipped from no to yes. |
| control | The unchanged answer is not flagged | POST /catalog-demo/caindrift/run | Same meaning: no false alarm. |
| attack | 'No, within 30 days' became 'Yes, any time': flagged | POST /catalog-demo/caindrift/run | A meaning flip a customer would act on. |
Each button runs the live comparison and checks one fact of it. Shared demo; rate limited per address.
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.