The scan's own causal reading from the sequence or structure, labelled as its own.
The definition the product enforces, read from the same code the methodology page renders.
The scan's own causal reading from the sequence or the structure, labelled as its own. This is a legitimate rung, not a failure: a well-labelled inference is worth more than a link dressed up as documented. Treat it as the scan's argument and test it.
How a cause-and-effect link is KNOWN, kept apart from how much it matters. A documented link can be weak; a strong link can be inferred.
Every grade here is assigned by the same rules on every scan, and the methodology page renders those rules from the code that enforces them.