One backbone. A family of products.
Everything Attaxr ships runs on the same event-driven core. Aquila ships today. The next products are in development on the same foundation.
Aquila
Our first product, and two systems on one backbone. The recon classifier runs the tools as services and scores findings while the scan is still running. Hackbot hunts: it takes the highest-value leads, probes them with an operator's judgment, and escalates what is real. Casper, our browser core, carries Hackbot through the pages that test their visitors, and Jev, our decision layer, keeps the budget on the leads that deserve it.
classifier
- Autonomous recon workflows
- Event-driven classification
- Millisecond triage and severity calls
- AI-generated reports
hackbot
- Chat-driven hunting sessions
- Probes the highest-ranked leads
- A browser core for pages that test for bots
The layers under the hunt
Two of them never appear in a dashboard: the browser the agents drive, and the decision layer that ranks their next move.
browser
Casper, the browser layer
The hackbot drives pages through Casper. Real browser profile, one identity per session, and answers that bot checks treat as a visitor's.
decision
Jev, the decision layer
Triage, severity, and the next investigation, decided in about a hundred milliseconds. The heavy models get the questions worth their time.
How an assessment runs
Four stages, one chain of events. Each stage reacts to the previous one automatically; the diagram is the architecture, not an aspiration.
recon
Recon runs as services
ProjectDiscovery tools stream subdomains, endpoints, and technologies as they are found. Nothing waits for a scan to finish.
classify
Classification starts mid-scan
Every discovery is an event. Findings are scored, correlated, and queued as leads while the sweep is still running.
hunt
Hackbot probes what matters
The highest-value leads get an operator's attention: focused probing, judgement about what is real, escalation when it pays off.
report
Reports are ready at the end
Ranked leads and written reports quote the evidence underneath them. The assessment closes itself; you review and act.
The platform underneath every product
Integration is not a feature we added; it is the architecture we started from.
events
One event backbone
Every finding and state change is an event on one backbone. Products you adopt join the same pipeline, so nothing is bolted on afterwards.
agents
One agent runtime
Specialist agents plan, call tools through the MCP catalog, and escalate. A decision layer, Jev, ranks where their budget goes. The same runtime drives every product we ship.
browser
One browser core
Every product that touches a page drives it through Casper: a real browser profile, one identity per session, built for targets that test their visitors.
evidence
One evidence store
Findings stay tied to what the tools observed. Reports quote the evidence; they do not paraphrase it.
The next products are already on the backbone.
They ship when they survive a real engagement, the same bar Aquila cleared. If your workflow should be one of them, tell us.
Tell us what to build firstBring autonomous security to your operation
Tell us about your targets and your workflow. We will show you what an assessment looks like when the agents run it end to end.





