Attack lab
Five real tool lists scanned live.
4 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 | |
|---|---|---|---|
| control | Ordinary add and send_email tools | POST /api/v1/mcp-tool-guard/scan | Nothing suspicious: passes. |
| attack | A description that grabs the SSH key | POST /api/v1/mcp-tool-guard/scan | Classic tool poisoning. |
| attack | A tool that changes another tool (BCC an outsider) | POST /api/v1/mcp-tool-guard/scan | Shadowing: one tool altering another. |
| attack | Invisible characters hiding an instruction | POST /api/v1/mcp-tool-guard/scan | Zero-width characters a person would never see. |
| attack | A 'helper' that asks for the .env file | POST /api/v1/mcp-tool-guard/scan | Credential grab. |
Each button scans one tool list on the live scanner. 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.