Demos
Watch Secronyx investigate real alerts
Six recorded runs from start to finish. In each one an alert fires in a tool teams already use, Secronyx gathers read-only evidence from inside the host, and an AI assistant works out what happened.
What the demos show
Evidence from inside the host, gathered safely
Monitoring tells you that something broke. The demos show the next step: an AI assistant looking inside the affected host to find out why, through read-only checks and without a shell.
Real alerts, real hosts
Each demo starts from an alert raised by the monitoring or security tool itself, on hosts set up for the run, and follows it through to the AI's conclusion.
Read-only and restricted
The AI reaches the hosts only through Secronyx's fixed, read-only checks. A label policy decides which hosts each caller may see, and each host writes every call to its own tamper-evident audit log.
Evidence you can check
The hosts sign their results. Each walkthrough verifies the evidence offline and shows the check failing once one character of a stored result is changed.
How each demo works
From alert to answer
An alert fires
A tool such as Grafana, Alertmanager or Falco raises an alert for one host.
It reaches Secronyx
The tool's own webhook sends it to the commander, or, in the AWS DevOps Agent demo, the agent calls the commander as its MCP server.
Secronyx gathers evidence
The host named in the alert runs fixed, read-only checks and signs each result. Hosts outside the caller's policy are not reached.
The AI works out what happened
The assistant reads the evidence, asks follow-up questions through Secronyx and reports what it found and what it could not check.
The demos
Pick the tool you run
Product names belong to their owners and are used only to say what each demo connects to. Secronyx is not affiliated with, sponsored or endorsed by them.