diff --git a/README.md b/README.md
index 6305b97..16e6204 100644
--- a/README.md
+++ b/README.md
@@ -554,7 +554,7 @@ Notice that nearly every feature below has at least subtle differences from Java
     <td>
       ✔ Unescaped <code>-</code> outside of range is literal in some contexts (different than JS rules in any mode)<br>
       ✔ Error for unescaped <code>[</code> that doesn't form nested class<br>
-      ✔ Allows leading unescaped <code>]</code><br>
+      ✔ Leading unescaped <code>]</code> OK<br>
       ✔ Fewer chars require escaping than JS<br>
     </td>
   </tr>
@@ -616,6 +616,7 @@ Notice that nearly every feature below has at least subtle differences from Java
     <td>
       ✔ Always "multiline"<br>
       ✔ Only <code>\n</code> as newline<br>
+      ✔ No match after string-terminating <code>\n</code><br>
     </td>
   </tr>
   <tr valign="top">
diff --git a/spec/match-assertion.spec.js b/spec/match-assertion.spec.js
index 0c61196..56ba10d 100644
--- a/spec/match-assertion.spec.js
+++ b/spec/match-assertion.spec.js
@@ -1,4 +1,4 @@
-import {toDetails} from '../dist/index.mjs';
+import {toDetails, toRegExp} from '../dist/index.mjs';
 import {r} from '../src/utils.js';
 import {matchers} from './helpers/matchers.js';
 
@@ -50,6 +50,12 @@ describe('Assertion', () => {
     it('should not match at positions other than the start of the string or after a line feed', () => {
       expect('ba').not.toFindMatch('^a');
     });
+
+    it('should not match after a string-terminating line feed', () => {
+      expect(''.match(toRegExp('^', {global: true}))).toHaveSize(1);
+      expect('\n'.match(toRegExp('^', {global: true}))).toHaveSize(1);
+      expect('\n\n'.match(toRegExp('^', {global: true}))).toHaveSize(2);
+    });
   });
 
   describe('string_end', () => {
diff --git a/src/generate.js b/src/generate.js
index 5707306..20fdd42 100644
--- a/src/generate.js
+++ b/src/generate.js
@@ -219,7 +219,8 @@ function genAssertion(node, _, gen) {
   if (kind === AstAssertionKinds.string_start) {
     return '^';
   }
-  // If a word boundary came through the transformer unaltered, that means `wordIsAscii` is enabled
+  // If a word boundary came through the transformer unaltered, that means `wordIsAscii` or
+  // `asciiWordBoundaries` is enabled
   if (kind === AstAssertionKinds.word_boundary) {
     return negate ? r`\B` : r`\b`;
   }
diff --git a/src/transform.js b/src/transform.js
index 8735387..72b0e23 100644
--- a/src/transform.js
+++ b/src/transform.js
@@ -136,8 +136,9 @@ const FirstPassVisitor = {
       // Onig's only line break char is line feed, unlike JS
       replaceWith(parseFragment(r`(?=\z|\n)`));
     } else if (kind === AstAssertionKinds.line_start) {
-      // Onig's only line break char is line feed, unlike JS
-      replaceWith(parseFragment(r`(?<=\A|\n)`));
+      // Onig's only line break char is line feed, unlike JS. Onig's `^` doesn't match after a
+      // string-terminating line feed
+      replaceWith(parseFragment(r`(?<=\A|\n(?!\z))`));
     } else if (kind === AstAssertionKinds.search_start) {
       if (!supportedGNodes.has(node) && !allowUnhandledGAnchors) {
         throw new Error(r`Uses "\G" in a way that's unsupported`);