Detonate everything. Trust nothing.

An AI-assisted sandbox that runs suspicious files, links and scripts in disposable, isolated environments, then tells your team exactly what they tried to do.

1:1

One sandbox
per sample

0 %

Implicit trust,
ever

Private beta
RFC 9116 disclosure

Platform

The far side
of your
network

Behaviour,
not signatures

Observed
Processes, files, network
Verdict
Explained, with evidence

Illustrative sample run

Every detonation records what a sample actually does (process trees, file writes, persistence attempts, outbound calls) and highlights the step that matters.

API-first
by design

Submit
REST and webhooks
Results
Structured JSON

Send samples straight from your SIEM, SOAR or CI pipeline and get verdicts back where your team already works.

How it works

Open it here.
Not there.

01

Submit

Drop in a file, a URL or a script, or send it straight from your pipeline.

02

Detonate

LunarVane spins up a fresh, isolated environment for each sample and records everything it does.

03

Contain

You get a plain-language verdict with the evidence behind it. Then the sandbox is destroyed.

Bring us
your worst file.

LunarVane is in private beta. Tell us what you'd want to detonate first and we'll get back to you.

Request early access

Found a security issue? security@lunarvane.com