diff --git a/eslint.config.js b/eslint.config.js
index 2538d3e4..aae7129b 100644
--- a/eslint.config.js
+++ b/eslint.config.js
@@ -33,16 +33,12 @@ export default tseslint.config(
       ],
 
       // unicorn rules:
-      'require-unicode-regexp': 'error',
       'unicorn/expiring-todo-comments': 'off',
       'unicorn/import-style': 'off',
       'unicorn/no-anonymous-default-export': 'off',
       'unicorn/no-array-reduce': 'off',
       'unicorn/no-nested-ternary': 'off',
       'unicorn/no-useless-undefined': 'off', // We use a lot of `return undefined` to satisfy the `consistent-return` rule.
-      'unicorn/prefer-at': 'off',
-      'unicorn/prefer-string-raw': 'off',
-      'unicorn/prefer-string-replace-all': 'off',
       'unicorn/prevent-abbreviations': 'off',
 
       // typescript-eslint rules:
@@ -112,6 +108,7 @@ export default tseslint.config(
       radix: 'error',
       'require-atomic-updates': 'error',
       'require-await': 'error',
+      'require-unicode-regexp': 'error',
       'spaced-comment': ['error', 'always', { markers: ['*', '!'] }],
       'sort-vars': 'error',
       yoda: 'error',
diff --git a/lib/markdown.ts b/lib/markdown.ts
index 92ae0d35..bffba98e 100644
--- a/lib/markdown.ts
+++ b/lib/markdown.ts
@@ -93,8 +93,8 @@ export function expectContentOrFail(
   // in case escaping is needed where the content is referenced.
   const hasContent =
     contents.includes(content) ||
-    contents.includes(content.replace(/"/gu, '\\"')) ||
-    contents.includes(content.replace(/'/gu, "\\'"));
+    contents.includes(content.replaceAll('"', String.raw`\"`)) ||
+    contents.includes(content.replaceAll("'", String.raw`\'`));
   if (hasContent !== expected) {
     console.error(
       `${docName} should ${
diff --git a/lib/rule-link.ts b/lib/rule-link.ts
index 5bbff61e..f08bbe9f 100644
--- a/lib/rule-link.ts
+++ b/lib/rule-link.ts
@@ -14,7 +14,7 @@ export function replaceRulePlaceholder(
   if (typeof pathOrUrl === 'function') {
     return pathOrUrl(ruleName);
   }
-  return pathOrUrl.replace(/\{name\}/gu, ruleName);
+  return pathOrUrl.replaceAll('{name}', ruleName);
 }
 
 /**
diff --git a/lib/rule-list.ts b/lib/rule-list.ts
index 4a0d09c0..78a567a0 100644
--- a/lib/rule-list.ts
+++ b/lib/rule-list.ts
@@ -365,8 +365,11 @@ function getRulesAndHeadersForSplit(
       // Turn ruleListSplit into a title.
       // E.g. meta.docs.requiresTypeChecking to "Requires Type Checking".
       const ruleListSplitParts = ruleListSplitItem.split('.');
-      const ruleListSplitFinalPart =
-        ruleListSplitParts[ruleListSplitParts.length - 1];
+      const ruleListSplitFinalPart = ruleListSplitParts.at(-1);
+      /* istanbul ignore next -- this shouldn't happen */
+      if (!ruleListSplitFinalPart) {
+        throw new Error(`Invalid ruleListSplit value "${ruleListSplitItem}".`);
+      }
       const ruleListSplitTitle = noCase(ruleListSplitFinalPart, {
         transform: (str) => capitalizeOnlyFirstLetter(str),
       });
diff --git a/lib/string.ts b/lib/string.ts
index a2d43bc5..e8210267 100644
--- a/lib/string.ts
+++ b/lib/string.ts
@@ -22,7 +22,9 @@ export function capitalizeOnlyFirstLetter(str: string) {
 }
 
 function sanitizeMarkdownTableCell(text: string): string {
-  return text.replace(/\|/gu, '\\|').replace(new RegExp(EOL, 'gu'), '<br/>');
+  return text
+    .replaceAll('|', String.raw`\|`)
+    .replaceAll(new RegExp(EOL, 'gu'), '<br/>');
 }
 
 export function sanitizeMarkdownTable(
diff --git a/test/lib/generate/option-postprocess-test.ts b/test/lib/generate/option-postprocess-test.ts
index 82ee92a6..63201181 100644
--- a/test/lib/generate/option-postprocess-test.ts
+++ b/test/lib/generate/option-postprocess-test.ts
@@ -49,7 +49,7 @@ describe('generate (postprocess option)', function () {
           [
             content,
             '',
-            `Located at ${relative('.', path).replace(/\\/gu, '/')}`, // Always use forward slashes in the path so the snapshot is right even when testing on Windows.
+            `Located at ${relative('.', path).replaceAll('\\', '/')}`, // Always use forward slashes in the path so the snapshot is right even when testing on Windows.
           ].join('\n'),
       });
       expect(readFileSync('README.md', 'utf8')).toMatchSnapshot();