Penetration Testing
In an OT environment, the test itself can be the incident. Scoping and safety constraints get as much attention here as the test methods do.
A test that affects the system it's meant to protect has failed regardless of what it finds
Penetration testing methods developed for enterprise IT do not transfer cleanly to operational technology. A technique that safely probes a web server can crash a programmable logic controller that has never been patched, was never designed to handle malformed traffic, and is presently running a safety-related function. In a nuclear or critical-infrastructure environment, that is not an acceptable risk to take in the name of finding one.
GSS scopes every engagement around what the environment can safely absorb before deciding what the test will attempt. That means excluding certain systems from active testing entirely, substituting passive analysis or a representative test-bed where live testing carries unacceptable risk, and agreeing in advance — with operations, not just with security — on what is and isn't in bounds.
The objective is the same as any penetration test: find what an adversary could actually exploit, under realistic constraints, and report it in terms operations and engineering can act on. The difference is that the path to that finding is built around the environment's tolerances first, not around a standard test playbook applied without regard for what it's being applied to.