AGENTPROOF
ATTACK AGENT

The Attack Agent looks for security problems in your website or web app by trying what an attacker would try. For each problem it finds, it saves the steps it took and what your app sent back.

The Attack Agent thinks like a hacker and tries to exploit your app with real techniques.

YOUR DATA
Can one customer open another customer's records by changing a number in the link?
ADMIN
Can a regular user reach pages or actions meant for admins?
SIGNED OUT
Can someone who is not signed in call requests meant for your users?
CHANGES
Can a user change a price, a role, or a setting they should not control?

The Attack Agent sends these requests to your app and checks what comes back.

Every problem it finds is saved with the exact request and response.

EXAMPLE FINDINGCustomer can open another customer's invoice
REQUESTGET /api/invoices/1043
Signed in as Customer A
RESPONSE200 OK
Invoice 1043 belongs to Customer B

Each finding includes the exact request, the response, and the steps to repeat it.

You choose which app gets tested and how often.

You choose the website or web app it may test. It tests again monthly, weekly, or daily, so new code gets checked too.

Problems go straight to the Defense Agent to be fixed.

Each finding goes straight to the Defense Agent, which writes the fix and runs the attack again.

See how fixes workSee plans

The animation above is a simulation. The Attack Agent is in development.