One environment per engagement
Each assessment runs in a clean environment built for it alone. Static analysis runs offline; network probes touch only in-scope targets during the agreed window.
The rules we work by. Each one is written into every statement of work.
We see only what the signed scope names, for as long as the engagement runs.
Each assessment runs in a clean environment built for it alone. Static analysis runs offline; network probes touch only in-scope targets during the agreed window.
Your code and configuration stay inside that environment, which is destroyed at close.
A manifest of tools, versions and inputs, with hashes of every output, so results can be reproduced.
We never publish client findings. Published statistics are aggregated, identify no client, and appear only where your contract allows.
Findings come from deterministic tools that give the same answer on every run. AI helps explain findings and draft guidance, an engineer reviews every line, and AI never creates a finding.
For legacy code where scanners are weak, an AI model flags candidate cryptography and an engineer confirms each one before it becomes a finding. Your code and configuration are never sent to public AI services: models run inside the engagement environment, or on government-authorized endpoints named in your agreement.
Served only over HTTPS, offering hybrid X25519MLKEM768 key exchange to browsers that support it.
No cookies, analytics, or third-party scripts and fonts. One self-hosted animation script, built from an open-source library and served from this domain; every page works without it.
A strict content security policy and a DNSSEC-signed domain.
SPF, DKIM and an enforced DMARC policy on every message from our domain.
Found a security issue in our website, email or tools? Tell us directly.