diff --git a/shared/tree-sitter-extractor/src/generator/mod.rs b/shared/tree-sitter-extractor/src/generator/mod.rs
index 8167dc2574e0..da13322fe60c 100644
--- a/shared/tree-sitter-extractor/src/generator/mod.rs
+++ b/shared/tree-sitter-extractor/src/generator/mod.rs
@@ -305,7 +305,18 @@ fn convert_nodes(
                 // type.
                 let members: Set<&str> = n_members
                     .iter()
-                    .map(|n| nodes.get(n).unwrap().dbscheme_name.as_str())
+                    .map(|n| {
+                        nodes
+                            .get(n)
+                            .unwrap_or_else(|| {
+                                panic!(
+                                    "union type '{}' references unknown member node type {:?}",
+                                    node.dbscheme_name, n
+                                )
+                            })
+                            .dbscheme_name
+                            .as_str()
+                    })
                     .collect();
                 entries.push(dbscheme::Entry::Union(dbscheme::Union {
                     name: &node.dbscheme_name,
diff --git a/shared/yeast-macros/src/parse.rs b/shared/yeast-macros/src/parse.rs
index 608c332d47e8..eb3b161b295a 100644
--- a/shared/yeast-macros/src/parse.rs
+++ b/shared/yeast-macros/src/parse.rs
@@ -411,7 +411,7 @@ fn parse_direct_node_inner(tokens: &mut Tokens, ctx: &Ident) -> Result<TokenStre
     // Named fields — compute each value into a temp, then reference it
     while peek_is_field(tokens) {
         let field_name = expect_ident(tokens, "expected field name")?;
-        let field_str = field_name.to_string();
+        let field_str = field_name.to_string().strip_prefix("r#").unwrap_or(&field_name.to_string()).to_string();
         expect_punct(tokens, ':', "expected `:` after field name")?;
         let temp = Ident::new(
             &format!("__field_{field_str}_{field_counter}"),
@@ -437,7 +437,11 @@ fn parse_direct_node_inner(tokens: &mut Tokens, ctx: &Ident) -> Result<TokenStre
                             .map(::std::convert::Into::<usize>::into)
                             .collect();
                     });
-                    field_args.push(quote! { (#field_str, #temp) });
+                    // An empty splice means the field is absent — skip it
+                    // entirely rather than emitting an empty named field.
+                    field_args.push(quote! {
+                        if !#temp.is_empty() { __fields.push((#field_str, #temp)); }
+                    });
                     continue;
                 }
             }
@@ -445,7 +449,7 @@ fn parse_direct_node_inner(tokens: &mut Tokens, ctx: &Ident) -> Result<TokenStre
 
         let value = parse_direct_node(tokens, ctx)?;
         stmts.push(quote! { let #temp: usize = #value; });
-        field_args.push(quote! { (#field_str, vec![#temp]) });
+        field_args.push(quote! { __fields.push((#field_str, vec![#temp])); });
     }
 
     // After all named fields, no other tokens are allowed.
@@ -461,7 +465,9 @@ fn parse_direct_node_inner(tokens: &mut Tokens, ctx: &Ident) -> Result<TokenStre
     Ok(quote! {
         {
             #(#stmts)*
-            #ctx.node(#kind_str, vec![#(#field_args),*])
+            let mut __fields: Vec<(&str, Vec<usize>)> = Vec::new();
+            #(#field_args)*
+            #ctx.node(#kind_str, __fields)
         }
     })
 }
@@ -475,6 +481,11 @@ fn parse_direct_list(tokens: &mut Tokens, ctx: &Ident) -> Result<Vec<TokenStream
             let group = expect_group(tokens, Delimiter::Parenthesis)?;
             let mut inner = group.stream().into_iter().peekable();
 
+            // Empty `()` represents an empty sequence — emit nothing.
+            if inner.peek().is_none() {
+                continue;
+            }
+
             // Regular node
             let node = parse_direct_node_inner(&mut inner, ctx)?;
             items.push(quote! { __nodes.push(#node); });
diff --git a/shared/yeast/src/captures.rs b/shared/yeast/src/captures.rs
index ef184cd9f69a..404d402a5016 100644
--- a/shared/yeast/src/captures.rs
+++ b/shared/yeast/src/captures.rs
@@ -63,16 +63,20 @@ impl Captures {
     }
 
     /// Apply a fallible function to every captured id (across all keys),
-    /// replacing each id with the result. Stops and returns the error on
-    /// the first failure.
+    /// replacing each id with the results. A function returning an empty
+    /// vector removes the capture; returning multiple ids splices them
+    /// into the capture's value list (suitable for `*`/`+` captures).
+    /// Stops and returns the error on the first failure.
     pub fn try_map_all_captures<E>(
         &mut self,
-        mut f: impl FnMut(Id) -> Result<Id, E>,
+        mut f: impl FnMut(Id) -> Result<Vec<Id>, E>,
     ) -> Result<(), E> {
         for ids in self.captures.values_mut() {
-            for id in ids {
-                *id = f(*id)?;
+            let mut new_ids = Vec::with_capacity(ids.len());
+            for &id in ids.iter() {
+                new_ids.extend(f(id)?);
             }
+            *ids = new_ids;
         }
         Ok(())
     }
diff --git a/shared/yeast/src/dump.rs b/shared/yeast/src/dump.rs
index 07ee134e058c..d046c192053d 100644
--- a/shared/yeast/src/dump.rs
+++ b/shared/yeast/src/dump.rs
@@ -273,6 +273,16 @@ fn dump_node(
         }
     }
 
+    // Check for required fields that are absent
+    if let Some((schema, _, _)) = type_check {
+        for (field_id, field_name) in schema.required_fields_for_kind(node.kind_name()) {
+            if !node.fields.contains_key(&field_id) {
+                let name = field_name.unwrap_or("child");
+                writeln!(out, "{prefix}  <-- ERROR: missing required field '{name}'").unwrap();
+            }
+        }
+    }
+
     // Unnamed children — skip unnamed tokens (keywords, punctuation)
     if let Some(children) = node.fields.get(&CHILD_FIELD) {
         let child_type_check = type_check.map(|(schema, _, _)| {
diff --git a/shared/yeast/src/lib.rs b/shared/yeast/src/lib.rs
index 8826c3d6c1e3..0717d27e4b8b 100644
--- a/shared/yeast/src/lib.rs
+++ b/shared/yeast/src/lib.rs
@@ -563,6 +563,15 @@ impl Node {
             NodeContent::DynamicString(s) => Some(s.to_string()),
         }
     }
+
+    /// Read the child ids stored under a given field, or an empty slice if
+    /// no such field is present on this node.
+    pub fn field_children(&self, field_id: FieldId) -> &[Id] {
+        self.fields
+            .get(&field_id)
+            .map(|v| v.as_slice())
+            .unwrap_or(&[])
+    }
 }
 
 /// The contents of a node is either a range in the original source file,
@@ -836,17 +845,9 @@ fn apply_one_shot_rules_inner(
                 // pattern root): re-analyzing it would match the same rule
                 // again indefinitely.
                 if captured_id == id {
-                    return Ok(captured_id);
-                }
-                let result =
-                    apply_one_shot_rules_inner(index, ast, captured_id, fresh, rewrite_depth + 1)?;
-                if result.len() != 1 {
-                    return Err(format!(
-                        "OneShot: recursion on captured node produced {} results, expected exactly 1",
-                        result.len()
-                    ));
+                    return Ok(vec![captured_id]);
                 }
-                Ok(result[0])
+                apply_one_shot_rules_inner(index, ast, captured_id, fresh, rewrite_depth + 1)
             })?;
             return Ok(rule.run_transform(ast, captures, id, fresh));
         }
diff --git a/shared/yeast/src/node_types_yaml.rs b/shared/yeast/src/node_types_yaml.rs
index eb191076be48..797f14cba720 100644
--- a/shared/yeast/src/node_types_yaml.rs
+++ b/shared/yeast/src/node_types_yaml.rs
@@ -314,6 +314,14 @@ fn apply_yaml_to_schema(
             node_types.sort_by(|a, b| a.kind.cmp(&b.kind).then(a.named.cmp(&b.named)));
             node_types.dedup_by(|a, b| a.kind == b.kind && a.named == b.named);
             schema.set_field_types(parent_kind, field_id, node_types);
+            schema.set_field_cardinality(
+                parent_kind,
+                field_id,
+                crate::schema::FieldCardinality {
+                    multiple: spec.multiple,
+                    required: spec.required,
+                },
+            );
         }
     }
 }
diff --git a/shared/yeast/src/query.rs b/shared/yeast/src/query.rs
index 01e5e22ad730..bcf0f7facab1 100644
--- a/shared/yeast/src/query.rs
+++ b/shared/yeast/src/query.rs
@@ -178,11 +178,15 @@ impl QueryListElem {
                     let Some(child) = remaining_children.next() else {
                         return Ok(false);
                     };
-                    if skip_unnamed {
-                        let node = ast.get_node(child).unwrap();
-                        if !node.is_named() {
-                            continue;
-                        }
+                    let node = ast.get_node(child).unwrap();
+                    // Skip tree-sitter `extras` (e.g. comments) during
+                    // positional matching: they are conceptually invisible
+                    // between siblings, mirroring tree-sitter query semantics.
+                    if node.is_extra() {
+                        continue;
+                    }
+                    if skip_unnamed && !node.is_named() {
+                        continue;
                     }
                     let snapshot = matches.clone();
                     if sub_query.do_match(ast, child, matches)? {
diff --git a/shared/yeast/src/schema.rs b/shared/yeast/src/schema.rs
index c832a57b23ad..bbd425f15a2c 100644
--- a/shared/yeast/src/schema.rs
+++ b/shared/yeast/src/schema.rs
@@ -8,6 +8,15 @@ pub struct NodeType {
     pub named: bool,
 }
 
+/// Multiplicity/optionality of a field declaration.
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub struct FieldCardinality {
+    /// Whether the field may hold more than one child.
+    pub multiple: bool,
+    /// Whether at least one child must be present.
+    pub required: bool,
+}
+
 /// A schema defining node kinds and field names for the output AST.
 /// Built from a node-types.yml file, independent of any tree-sitter grammar.
 ///
@@ -32,6 +41,7 @@ pub struct Schema {
     kind_names: BTreeMap<KindId, &'static str>,
     next_kind_id: KindId,
     field_types: BTreeMap<(String, FieldId), Vec<NodeType>>,
+    field_cardinalities: BTreeMap<(String, FieldId), FieldCardinality>,
     supertypes: BTreeMap<String, Vec<NodeType>>,
 }
 
@@ -52,6 +62,7 @@ impl Schema {
             kind_names: BTreeMap::new(),
             next_kind_id: 1, // 0 is reserved
             field_types: BTreeMap::new(),
+            field_cardinalities: BTreeMap::new(),
             supertypes: BTreeMap::new(),
         }
     }
@@ -196,6 +207,42 @@ impl Schema {
             .get(&(parent_kind.to_string(), field_id))
     }
 
+    pub fn set_field_cardinality(
+        &mut self,
+        parent_kind: &str,
+        field_id: FieldId,
+        cardinality: FieldCardinality,
+    ) {
+        self.field_cardinalities
+            .insert((parent_kind.to_string(), field_id), cardinality);
+    }
+
+    /// Returns the declared cardinality for a field, if known.
+    pub fn field_cardinality(
+        &self,
+        parent_kind: &str,
+        field_id: FieldId,
+    ) -> Option<FieldCardinality> {
+        self.field_cardinalities
+            .get(&(parent_kind.to_string(), field_id))
+            .copied()
+    }
+
+    /// Returns an iterator over all `(field_id, field_name)` pairs that are
+    /// declared as required (`required: true`) for the given `parent_kind`.
+    pub fn required_fields_for_kind<'a>(
+        &'a self,
+        parent_kind: &'a str,
+    ) -> impl Iterator<Item = (FieldId, Option<&'static str>)> + 'a {
+        self.field_cardinalities
+            .iter()
+            .filter(move |((kind, _), card)| kind == parent_kind && card.required)
+            .map(move |((_, field_id), _)| {
+                let name = self.field_name_for_id(*field_id);
+                (*field_id, name)
+            })
+    }
+
     pub fn set_supertype_members(&mut self, supertype: &str, node_types: Vec<NodeType>) {
         self.supertypes.insert(supertype.to_string(), node_types);
     }
diff --git a/shared/yeast/tests/test.rs b/shared/yeast/tests/test.rs
index 7f4db1541836..7645f3776f87 100644
--- a/shared/yeast/tests/test.rs
+++ b/shared/yeast/tests/test.rs
@@ -274,6 +274,44 @@ fn test_query_no_match() {
     assert!(!matched);
 }
 
+#[test]
+fn test_query_skips_extras_in_positional_match() {
+    // Regression test: positional wildcards `(_)` must not bind to
+    // tree-sitter `extras` (e.g. comments) during forward-scan; extras
+    // are conceptually invisible between siblings, matching tree-sitter
+    // query semantics. Without this, a later rule that translates a
+    // captured comment to nothing (a common idiom, e.g.
+    // `(comment) => ()` in Swift) leaves the capture's match-list empty
+    // and causes the transform to fail with "Variable X has 0 matches".
+    let runner = Runner::new(tree_sitter_ruby::LANGUAGE.into(), &[]);
+    let ast = runner.run("[1, # comment\n2]").unwrap();
+
+    // Navigate to the `array` node: program -> array.
+    let mut cursor = AstCursor::new(&ast);
+    cursor.goto_first_child();
+    let array_id = cursor.node_id();
+    assert_eq!(ast.get_node(array_id).unwrap().kind(), "array");
+
+    // Two positional wildcards should bind to the two integers, skipping
+    // the comment that sits between them.
+    let query = yeast::query!((array (_) @a (_) @b));
+    let mut captures = yeast::captures::Captures::new();
+    let matched = query.do_match(&ast, array_id, &mut captures).unwrap();
+    assert!(matched);
+    assert_eq!(
+        ast.get_node(captures.get_var("a").unwrap())
+            .unwrap()
+            .kind(),
+        "integer"
+    );
+    assert_eq!(
+        ast.get_node(captures.get_var("b").unwrap())
+            .unwrap()
+            .kind(),
+        "integer"
+    );
+}
+
 #[test]
 fn test_reachable_nodes_excludes_orphaned_rewrite_nodes() {
     let lang: tree_sitter::Language = tree_sitter_ruby::LANGUAGE.into();
diff --git a/unified/extractor/tests/corpus_tests.rs b/unified/extractor/tests/corpus_tests.rs
index ac93dd586ec1..0f1057a8e5b9 100644
--- a/unified/extractor/tests/corpus_tests.rs
+++ b/unified/extractor/tests/corpus_tests.rs
@@ -26,6 +26,13 @@ fn is_header_rule(line: &str) -> bool {
     trimmed.len() >= 3 && trimmed.chars().all(|c| c == '=')
 }
 
+fn is_next_case_header(lines: &[&str], i: usize) -> bool {
+    is_header_rule(lines[i])
+        && i + 2 < lines.len()
+        && !lines[i + 1].trim().is_empty()
+        && is_header_rule(lines[i + 2])
+}
+
 fn parse_corpus(content: &str) -> Vec<CorpusCase> {
     let lines: Vec<&str> = content.lines().collect();
     let mut i = 0;
@@ -58,48 +65,51 @@ fn parse_corpus(content: &str) -> Vec<CorpusCase> {
 
         let input_start = i;
         while i < lines.len() && lines[i].trim() != "---" {
+            if is_next_case_header(&lines, i) {
+                break;
+            }
             i += 1;
         }
-        assert!(i < lines.len(), "Missing --- separator for case {name}");
         let input = lines[input_start..i].join("\n").trim_end().to_string();
-        i += 1;
+        let raw;
+        let expected;
+        if i >= lines.len() || lines[i].trim() != "---" {
+            // No `---` separator before next case (or EOF). Treat the
+            // remaining sections as empty.
+            raw = String::new();
+            expected = String::new();
+        } else {
+            i += 1;
 
-        // Raw tree-sitter parse section. New-format files have a second
-        // `---` separator between the raw tree and the mapped AST. Legacy
-        // files (with only one separator) have no raw section — in that
-        // case `raw` stays empty and update mode will populate it.
-        let raw_start = i;
-        let mut next_sep = i;
-        while next_sep < lines.len() && lines[next_sep].trim() != "---" {
-            if is_header_rule(lines[next_sep])
-                && next_sep + 2 < lines.len()
-                && !lines[next_sep + 1].trim().is_empty()
-                && is_header_rule(lines[next_sep + 2])
-            {
-                break;
+            // Raw tree-sitter parse section. New-format files have a second
+            // `---` separator between the raw tree and the mapped AST. Legacy
+            // files (with only one separator) have no raw section — in that
+            // case `raw` stays empty and update mode will populate it.
+            let raw_start = i;
+            let mut next_sep = i;
+            while next_sep < lines.len() && lines[next_sep].trim() != "---" {
+                if is_next_case_header(&lines, next_sep) {
+                    break;
+                }
+                next_sep += 1;
             }
-            next_sep += 1;
-        }
-        let raw = if next_sep < lines.len() && lines[next_sep].trim() == "---" {
-            let raw_text = lines[raw_start..next_sep].join("\n").trim().to_string();
-            i = next_sep + 1;
-            raw_text
-        } else {
-            String::new()
-        };
+            raw = if next_sep < lines.len() && lines[next_sep].trim() == "---" {
+                let raw_text = lines[raw_start..next_sep].join("\n").trim().to_string();
+                i = next_sep + 1;
+                raw_text
+            } else {
+                String::new()
+            };
 
-        let expected_start = i;
-        while i < lines.len() {
-            if is_header_rule(lines[i])
-                && i + 2 < lines.len()
-                && !lines[i + 1].trim().is_empty()
-                && is_header_rule(lines[i + 2])
-            {
-                break;
+            let expected_start = i;
+            while i < lines.len() {
+                if is_next_case_header(&lines, i) {
+                    break;
+                }
+                i += 1;
             }
-            i += 1;
+            expected = lines[expected_start..i].join("\n").trim().to_string();
         }
-        let expected = lines[expected_start..i].join("\n").trim().to_string();
 
         cases.push(CorpusCase {
             name,