Attack lab

Six live drills, each a different memory attack.

6 attacks 0 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	A planted 'already approved' note	`POST /api/v1/memory-firewall/verify-drill`	Tainted, dropped from recall, can never become a permission.
attack	An agent tries to write a policy into memory	`POST /api/v1/memory-firewall/verify-drill`	Agents cannot write rules, instructions or credentials.
attack	A live API key written into memory	`POST /api/v1/memory-firewall/verify-drill`	13 known secret formats refused, even split by invisible characters.
attack	A revoked permission comes back from memory	`POST /api/v1/memory-firewall/verify-drill`	Revoked stays revoked.
attack	One customer reads another customer's memory	`POST /api/v1/memory-firewall/verify-drill`	Tenants are isolated.
attack	Summarising memories to gain authority	`POST /api/v1/memory-firewall/verify-drill`	Summaries can only narrow, never add authority.

Each button runs a full drill on the live firewall with a signed receipt. 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.