diff --git a/.github/actions/prose-gate/prose_lint.py b/.github/actions/prose-gate/prose_lint.py
index 4e2896d18..d6802fd79 100755
--- a/.github/actions/prose-gate/prose_lint.py
+++ b/.github/actions/prose-gate/prose_lint.py
@@ -2371,8 +2371,9 @@ def check_file(path: Path, rules: set[str], root: Path | None = None) -> list[tu
         # Both prose rules are Markdown-only until a comment can be told from code.
         # A shell script carries 78 statement separators that are not prose at all.
         if path.suffix == ".md":
+            unquoted = QUOTE_PREFIX.sub(lambda m: " " * len(m.group()), prose, count=1)
             if "semicolon" in rules:
-                for span in list_spans(prose):
+                for span in list_spans(unquoted):
                     # The sentence is the unit, since the list an exemption protects lives in one.
                     # Judged over a whole bullet, one colon exempted every semicolon after it.
                     for sentence in sentences(span):
@@ -2388,7 +2389,6 @@ def check_file(path: Path, rules: set[str], root: Path | None = None) -> list[tu
                                 (i, "semicolon", "semicolon in prose -> a comma or two sentences")
                             )
             if "dash" in rules:
-                unquoted = QUOTE_PREFIX.sub(lambda m: " " * len(m.group()), prose, count=1)
                 skip = LABEL_DASH.match(unquoted)
                 for m in DASH.finditer(unquoted):
                     if skip and m.start() < skip.end():
diff --git a/tests/test_prose_lint.py b/tests/test_prose_lint.py
index 809686bb7..fd46df6c5 100755
--- a/tests/test_prose_lint.py
+++ b/tests/test_prose_lint.py
@@ -519,6 +519,36 @@ def test_a_bullet_label_colon_inside_the_emphasis_is_the_same_opener(self) -> No
             ),
         )
 
+    def test_a_blockquoted_bullet_label_colon_does_not_announce_a_list(self) -> None:
+        """A callout's `>` is structure, so its bullet is judged as the same bullet unquoted."""
+        for text in (
+            "- **Label**: one, two; three\n",
+            "> - **Label**: one, two; three\n",
+            ">- **Label**: one, two; three\n",
+            "> > - **Label**: one, two; three\n",
+            "> 1. **Label:** one, two; three\n",
+        ):
+            with self.subTest(text=text.strip()):
+                self.assertEqual(["semicolon"], self.kinds(text, {"semicolon"}))
+
+    def test_a_blockquoted_line_keeps_the_exemptions_it_has_unquoted(self) -> None:
+        """Only the prefix is blanked, so a list in a callout keeps its semicolons."""
+        self.assertEqual(
+            [], self.kinds("> Inputs: a, b, and c; outputs: d and e.\n", {"semicolon"})
+        )
+
+    def test_a_blockquoted_table_row_judges_each_cell_alone(self) -> None:
+        """A row inside a callout is still a record of fields rather than one sentence.
+
+        Read whole, the first cell's colon and the last cell's comma made the row a list.
+        """
+        for text in (
+            "| S1: x | it runs; it gates | D1, D2 |\n",
+            "> | S1: x | it runs; it gates | D1, D2 |\n",
+        ):
+            with self.subTest(text=text.strip()):
+                self.assertEqual(["semicolon"], self.kinds(text, {"semicolon"}))
+
     def test_an_abbreviation_does_not_end_a_sentence(self) -> None:
         """Splitting at `e.g.` cuts a list in half and flags the separator the exemption protects."""
         self.assertEqual(