Method
NIIS 9 treats method as part of the product. The sequence below is the default. It bends for the system, not for a scanner license. Scroll it and watch the constellation lock.
Work starts with what would actually hurt (a credential, a clinical workflow, a fleet key) and who is in a position to hurt it. Authorization boundaries are written down before any radio is touched. If the adversary is “anyone on the platform,” that is a different engagement than “a nation-state with a van.”
Interfaces, identifiers, pairing state, firmware trains, mobile OS services, backend contracts, and the public footprint go on one diagram. Hidden radios in accessories are first-class nodes. So are vendor SDKs.
STRIDE and LINDDUN questions are applied to each boundary. The output is a set of hypotheses: this pairing can be replayed; this MAC rotation is cosmetic; this attestation does not bind the radio. Hypotheses are cheaper to debate than to discover at the end of a test window.
Testing is manual and targeted. It exists to confirm, refute, or sharpen hypotheses. There is no coverage trophy for running every plugin. Where a demonstration of impact is needed, it is bounded by the authorization that was framed in step one.
Reports rank by operational impact, not by the loudness of a CVSS string. Each finding points at a control, a design choice, or an assumption. Remediation is written so it can be scheduled. Retest is part of the same arc, not a sequel pitch.
Refusals
