diff --git a/custom_components/localthings/registry/capabilities/operational.py b/custom_components/localthings/registry/capabilities/operational.py
index 440cc8d..e54db0c 100644
--- a/custom_components/localthings/registry/capabilities/operational.py
+++ b/custom_components/localthings/registry/capabilities/operational.py
@@ -67,11 +67,14 @@ def _new_cycle_running(rep):
     early once a new cycle is genuinely running.
 
     Gated on `state == 'active'`, unlike _just_finished's arm condition
-    above: issue #358's dryer replays a running stage ('Drying') for a few
-    seconds after Finish while `state` already reads idle, and a bypass
-    keyed on the progress code alone read that tail as a new cycle and
-    republished it. Releasing late costs nothing -- an unreleased hold
-    still expires on its own -- so this side takes the stronger signal."""
+    above: issue #358's dryer resets `progress` to its course's first
+    stage ('Drying') in the same moment `state` goes idle, ~4s before
+    settling to 'None' -- confirmed by the reporter's machine_state
+    history, which flips to idle on the exact second progress reads
+    'Drying', in both captured cycles. A bypass keyed on the progress
+    code alone read that as a new cycle and republished it. Releasing
+    late costs nothing -- an unreleased hold still expires on its own --
+    so this side takes the stronger signal."""
     v = rep.get("x.com.samsung.da.progress")
     return _state_is_active(rep) and v is not None and v not in ("None", "Finish")
 
diff --git a/tests/test_sensor_sticky.py b/tests/test_sensor_sticky.py
index 962f0eb..0627a9b 100644
--- a/tests/test_sensor_sticky.py
+++ b/tests/test_sensor_sticky.py
@@ -166,12 +166,13 @@ def test_a_new_cycle_starting_overrides_the_hold():
 
 
 def test_a_running_stage_after_finish_does_not_break_the_hold():
-    """Issue #358: the reporting dryer replays a running stage after
-    Finish -- observed twice, identically, as Cooling -> +60s Finish ->
-    +24s 'Drying' -> +4s settled, with `state` already idle throughout the
-    tail. The hold must survive it, so the cycle still reads Drying,
-    Cooling, Finish, Idle rather than the reported Drying, Cooling,
-    Finish, Drying, Idle."""
+    """Issue #358: the reporting dryer resets `progress` to its course's
+    first stage in the same moment `state` goes idle -- observed twice,
+    identically, as Cooling -> +60s Finish -> +24s 'Drying' -> +4s
+    settled, with the reporter's machine_state history flipping to idle on
+    the exact second progress reads 'Drying'. The hold must survive that,
+    so the cycle still reads Drying, Cooling, Finish, Idle rather than the
+    reported Drying, Cooling, Finish, Drying, Idle."""
     desc = replace(_PROGRESS_DESC, sticky_seconds=0.2)
     sensor, coordinator = _sensor(desc)