Every attack starts with your objective.
- Deployment
- A fresh Docker container for each campaign.
- Starting position
- A Linux host with access to your authorized network.
New access changes the next attack.
A credential recovered from a readable configuration file can give the agent access to another server.
The goal can be Active Directory domain control, administrator access to a server, or a permitted read of sensitive data. You define the access that matters and the systems the agent may test.
Follow the access behind the result.
The report connects the access gained toward your goal to the steps that enabled it.
Inspect how an exposed credential or permission contributed to the attack. Subscribers can read target identities, captured evidence and remediation advice beside each access step.
How to read an attack reportTest the fix with a fresh campaign.
After changing a permission or closing an exposure, run a new campaign against the affected objective and scope. Review the access gained from that starting position.
You set the boundaries.
Authorize the network scope and review exclusions and reversible changes before deployment. You can request a stop from the campaign page.
Run the campaign’s Docker command on a dedicated Linux host with rootful Docker. Target systems do not need a preinstalled agent.
Read the experiments.
Our published research documents the starting conditions, attack paths and access reached. Each experiment keeps its method and results together.
- AutoAttack vs Nessus
- What a vulnerability scan reported and what an attack demonstrated from the same starting position.
- AutoAttack and NodeZero
- Repeated AutoAttack runs alongside Horizon3’s separately published experiment, with the differences between the tests made explicit.
Put your objective to the test.
Run a campaign free and read the attack narrative with masked target names. A subscription reveals the identities, captured evidence and remediation advice.
ContactDiscuss your assessment.
Ask about an objective, deployment requirements or pricing.
hello@autoattack.aiReport vulnerabilities in AutoAttack to security@autoattack.ai.