Scope
One host and one external ingress, named in a written authorization. We do not see your source, your cloud console, or your internal network.
Observ runs a contained agent against one host you authorize. You get a report of what it found, with evidence. The clone is destroyed when the run ends.
A fixed scope keeps the test honest and the price flat. No hourly table, no growing surface area, no surprise invoice.
One host and one external ingress, named in a written authorization. We do not see your source, your cloud console, or your internal network.
An agent reconnoiters the ingress, investigates what it finds, and records only what it can prove. A finding without evidence does not count.
A ranked list of findings with severity, location, and the request that demonstrates each one. Delivered as a document you keep.
Deposit reserves a window. A signed scope starts the run.
The agent starts with one address and nothing else. It cannot see the runner that scores it, the answer key, or anything else on the machine.
You sign the host, the window, and the rules.
We boot a disposable clone of the target. Nothing else is touched.
The agent maps what is exposed on the single ingress.
It investigates within policy and attaches evidence to every finding.
Findings are normalized, scored, and written up.
The clone and the run environment are torn down. Nothing is left running.
The only path between them is the ingress you approved.
The gate is written authorization. If we cannot see it, we do not start.
Every entry in the report carries the same fields, so your team can reproduce it and act on it. No evidence, no finding.
Exposure
host.example.com:443
HTTP 200 on the admin path with no authentication challenge. Raw response captured.
Confirmed. Report only. No action taken against the host.
One host, one run, $3,995. Tell us the host and the window, and we will confirm scope.
Deposit reserves a window. Written scope starts the run.