R's treering holds 7,980 yearly tree-ring widths, 6000 BC to 1979. There is no documented step in it, so every changepoint a detector reports is a false alarm. The default mean model reports 38.
The lag-1 autocorrelation is only about 0.16, so a warning keyed to high autocorrelation stays quiet. A separate check compares block means: they vary about 6 times as much as white noise would allow. The tool already prints a "slow drift or a cycle" hint when that happens. The project's to-do list still said this series "gets no hint", which was out of date.
I confirmed the hint fires and added a test to quiet-series.test.mjs so it can't quietly disappear. Still open: the hint cannot tell a drift from a cycle.
What this does not show: that the hint works on other series. It is one real series, one warning.