One self-explaining diagnosis. id is the stable rule key; scenario names what
the runtime is doing wrong in plain language; advice is the fix; evidence is
the human-readable metric values that tripped the rule, so a trace explains
itself to a human or an agent without re-deriving the thresholds.
One self-explaining diagnosis. id is the stable rule key; scenario names what the runtime is doing wrong in plain language; advice is the fix; evidence is the human-readable metric values that tripped the rule, so a trace explains itself to a human or an agent without re-deriving the thresholds.