CAIN-42 MCPGateAttack lab
Six real one-sentence descriptions, classified live. Results are deterministic: the same sentence always gets the same tier.
4 attacks 2 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 | |
|---|---|---|---|
| control | An email spam filter | POST /catalog-demo/legal-auditor/run | The Commission's own example of minimal-risk AI. |
| control | A customer-service chatbot | POST /catalog-demo/legal-auditor/run | People must be told they are talking to an AI. |
| attack | Ranks job applicants and filters CVs | POST /catalog-demo/legal-auditor/run | Employment decisions are a named high-risk use. |
| attack | Scores exams to decide university admission | POST /catalog-demo/legal-auditor/run | Education admission is a named high-risk use. |
| attack | Predicts who will commit a crime from personality | POST /catalog-demo/legal-auditor/run | Predicting crime from profiling alone is banned. |
| attack | Live facial recognition in streets for police | POST /catalog-demo/legal-auditor/run | Real-time public biometric identification by police is banned except in narrow cases. |
Each button classifies one sentence on the live service. 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.