What is written in my code?
It is the code family: your code, read line by line.
Our products answer two questions, and each one has its own family.
Two questions
It is the code family: your code, read line by line.
It is the server family: the environment around it — who gets in, what is listening, what starts on its own, what was left open by an old configuration, what the backup takes.
They are two different readings, and each one answers its own question. The targets are independent too: the company website and a mobile app may have no connection at all, and each one comes in through the product that matches it.
The code family
Your code, read line by line. There are three products, and what changes between them is where your code is when the audit reads it.
the code is on your machine
Your project folder, read where it already is.
audit on the server where the system runs
The same code, read on the server where the system is up. What the system does in production is part of the reading.
the audit runs on our infrastructure
The same code, audited on our infrastructure. You send the address of where it is and grant read access; the report arrives in your inbox with the data destruction certificate.
The AI account, and where it comes in. Where the audit is executed in your environment, it runs on the Claude Code of your own account: the AI credit is yours, billed by Anthropic straight to your account. It runs to the end with Claude Max, Team or Enterprise, or with API credits. In Code Prime the processing is included in the price.
How to install and how to pay for the AI →The server family
The infrastructure your system runs on. There are three products, and what changes between them is where you look from, and how many domains the server hosts.
your site, with access to your hosting
A domain of yours on shared hosting, audited from the inside: you paste our public key into your panel and the audit runs on our server.
your own server with root, hosting one domain
The infrastructure of the server that hosts a domain of yours: the firewall, remote access, disk space, the ports, the services that start, the permissions, the updates, the scheduled tasks and what the backup takes. The code running there belongs to the other family.
sudo.the server that hosts several domains — you are the owner
The infrastructure reading of the server, done once and holding for the whole machine, plus the application level on every domain you declare: the vhost, the application, the certificate and the route.
On Host, MultiHost and Share
The audit reads the version of every piece that is live — on Host and MultiHost, the operating system, the web server, the language, the database and remote access; on Share, whatever serves your account — and answers three questions about each one:
The third one is the one you settle today: it asks for no migration and no waiting on a vendor, only applying what has already shipped. On Host and MultiHost the report states how many packages are upgradable, how many of those are security updates, which ones are on the exposed surface, and whether the server is waiting on a restart to finish what has already been installed.
On Share the machine belongs to the provider, and applying the update is not in your hands. What the report delivers there is the number actually serving your site, and the distance between it and what the panel offers — because the version answering on the command line is often not the one answering on the web.
The version is read where it is live: from the package manager, where the server is yours, and from the configuration that serves the site, on Share. The audit does not execute your system to ask it for the version.
The add-on
You can add the WordPress site analysis to Share, to Code Prime, to Host and to MultiHost.
Every plugin and every theme is compared against the official package, file by file. That is how a tampered plugin on the latest version shows up — the kind that passes any check made by version number. It also covers signs of a planted file, permissions, the isolation of your account on the server, and credentials stored in a public folder.
Where it is executed depends on the model. On Share and Code Prime, on our infrastructure: you paste the public key we send into your hosting panel, and you can revoke it whenever you want. On Host and MultiHost, inside your own environment, with the access you already use — there we neither ask for nor receive any access at all. The domain is the one of the WordPress site being audited, and the amount appears as a separate line in the quote.
The two levels
One agent searches, and that same agent judges each finding against the code.
Several agents search in parallel, and other agents judge, without having taken part in the discovery.
Deep reaches further and costs more, and it pays off where there are boundaries: separate subsystems, client and server, a binary talking to the host, code talking to the database. We measured both cases: in a system with many boundaries Deep found 4.5× more; in a flat folder, 1.23×.
When in doubt, ask for both prices: the form has the option both — I want both prices to decide, which is the screen's default. With both numbers in hand, and knowing your system, the choice is yours.
In Share the level is ONE. In the other five products you choose the level, or ask for both prices.
What comes with any of them
In every product, each finding that enters the report comes with file, line and the trail: the chain of lines from the entry point to the point that executes.
And the refuted ones come too, in a section of their own, with what was suspected and the reason it fell. They are there on purpose: a refutation can be wrong as well, and with the reason written down your team can disagree with us.
See the report, field by field →Pick the product, and see the price: in the server family the amount appears on screen; in the code family it arrives in your inbox after the measurement.