Attack lab

One envelope for a low-risk support agent (search and read only, $100 total, low consequences). Six real paths checked live:

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	Searches the knowledge base	`POST /fabric/argus/envelope/evaluate`	Inside every limit.
control	Reads a file, then searches	`POST /fabric/argus/envelope/evaluate`	Both tools allowed.
attack	Searches, then attempts a $5,000 wire	`POST /fabric/argus/envelope/evaluate`	Forbidden tool, critical consequence.
attack	Uses a tool not on the list (send_email)	`POST /fabric/argus/envelope/evaluate`	Not permitted, even though it is 'just' email.
attack	Two $60 steps (total $120 > $100)	`POST /fabric/argus/envelope/evaluate`	Amounts add up across the path.
attack	Reads payroll (high-consequence data)	`POST /fabric/argus/envelope/evaluate`	Allowed tool, but the consequence is above the envelope.

Each button evaluates one path on the live engine. 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.