Role file

Dupery Milkman: alignment changes after the twenty-fourth hour

The short answer

Milkman is normally a good Meddler, but the Demo description says it becomes evil after the twenty-fourth hour. Preserve the time of an alignment clue: an earlier good result need not describe its later state.

Category
Meddler
Alignment
Normally Good
Ability
Becomes evil after the twenty-fourth hour in the Demo rule.

Uses the Demo v0.8.2 wording “after the 24th hour”; it does not claim an at-exactly-24-hour trigger in every build.

Time changes alignment, not the category label

Milkman starts from the normal Meddler/good grouping. The timed rule changes its alignment to evil after the twenty-fourth hour. That is why a reference’s normal-alignment label is not enough to decide its current status late in a case.

Keep role category, current alignment and investigation time as separate notes. An alignment change is not by itself a new disguise or a taint effect.

An old clue can stop being current

In an illustrative case, a reliable alignment clue obtained at hour 20 says a target is good. Later, after the Milkman threshold, another reliable check can describe it as evil. If the target is Milkman and the rule applies, this is a state change rather than a necessary contradiction between truthful sources.

Do not erase the earlier note or call the old source a liar just because the current state differs. Record “good at hour 20” and separately establish what is true now.

Milkman before changeNormally good MeddlerTime advancesafter the 24th hourCurrent alignment: evilCategory: Meddler
Illustrates the Demo’s “after” wording, not a claim about an exact frame at hour 24. An earlier clue and a current clue can describe different states.

Check the clock before using an older conclusion

Read the actual trigger in your build before making a time-sensitive decision. This page keeps the Demo’s “after” wording; it does not invent a precise frame, action order or inclusive boundary for a different version.

If the case has crossed a relevant time threshold, revisit constraints that depend on the target’s alignment, including later evil-counting information. Do not assume an already displayed statement automatically refreshes: check when its ability obtains or updates information.

Use a fresh, applicable clue for the current state, or preserve the old clue as historical information rather than a current clearance.

Further reading