diff --git a/.cspell.json b/.cspell.json
index c7bea3bd6aa7..3e9dd8c3257a 100644
--- a/.cspell.json
+++ b/.cspell.json
@@ -53,6 +53,6 @@
     "*.min.*",
     "jest/vendor"
   ],
-  "ignoreRegExpList": ["Email", "Urls", "#[\\w-]*"],
+  "ignoreRegExpList": ["Email", "Urls", "#[\\w-]*", "\\u\\{[0-9A-F]{4}\\}"],
   "enableFiletypes": ["mdx"]
 }
diff --git a/packages/docusaurus-utils/src/__tests__/markdownUtils.test.ts b/packages/docusaurus-utils/src/__tests__/markdownUtils.test.ts
index a81e8cfff103..e81622499d8b 100644
--- a/packages/docusaurus-utils/src/__tests__/markdownUtils.test.ts
+++ b/packages/docusaurus-utils/src/__tests__/markdownUtils.test.ts
@@ -197,6 +197,26 @@ describe('createExcerpt', () => {
         `),
     ).toBe('Lorem ipsum dolor sit amet, consectetur adipiscing elit.');
   });
+
+  it('creates excerpt with XML tag inside inline code', () => {
+    expect(
+      createExcerpt(dedent`
+          # Markdown Regular Title
+
+          This paragraph includes a link to the \`<metadata>\` documentation.
+        `),
+    ).toBe('This paragraph includes a link to the <metadata> documentation.');
+  });
+
+  it('creates excerpt with XML tag inside inline code with hyperlink', () => {
+    expect(
+      createExcerpt(dedent`
+          # Markdown Regular Title
+
+          This paragraph includes a link to the [\`<metadata>\`](https://developer.mozilla.org/en-US/docs/Web/SVG/Element/metadata) documentation.
+        `),
+    ).toBe('This paragraph includes a link to the <metadata> documentation.');
+  });
 });
 
 describe('parseMarkdownContentTitle', () => {
diff --git a/packages/docusaurus-utils/src/markdownUtils.ts b/packages/docusaurus-utils/src/markdownUtils.ts
index 216d0122e7ad..9626e14ec4d1 100644
--- a/packages/docusaurus-utils/src/markdownUtils.ts
+++ b/packages/docusaurus-utils/src/markdownUtils.ts
@@ -16,6 +16,12 @@ import type {
 // server-side when we infer metadata like `title` and `description` from the
 // content. Most parsing is still done in MDX through the mdx-loader.
 
+/**
+ * Characters that have markup behavior for the Markdown/MDX renderer to
+ * interpret.
+ */
+const MARKUP_CHARS = ['_', ':', '*', '<', '>', '~', '!', '[', ']'];
+
 /**
  * Hacky temporary escape hatch for Crowdin bad MDX support
  * See https://docusaurus.io/docs/i18n/crowdin#mdx
@@ -143,6 +149,12 @@ export function createExcerpt(fileString: string): string | undefined {
     }
 
     const cleanedLine = fileLine
+      // Unwrap inline code.
+      .replace(/`(?<text>.+?)`/g, (_, text) => {
+        return MARKUP_CHARS.reduce((acc, val) => {
+          return acc.replaceAll(val, `\u{FFFE}${val.codePointAt(0)}\u{FFFF}`);
+        }, text);
+      })
       // Remove HTML tags.
       .replace(/<[^>]*>/g, '')
       // Remove Title headers
@@ -159,8 +171,6 @@ export function createExcerpt(fileString: string): string | undefined {
       .replace(/\[\^.+?\](?:: .*$)?/g, '')
       // Remove inline links.
       .replace(/\[(?<alt>.*?)\][[(].*?[\])]/g, '$1')
-      // Remove inline code.
-      .replace(/`(?<text>.+?)`/g, '$1')
       // Remove blockquotes.
       .replace(/^\s{0,3}>\s?/g, '')
       // Remove admonition definition.
@@ -172,7 +182,9 @@ export function createExcerpt(fileString: string): string | undefined {
       .trim();
 
     if (cleanedLine) {
-      return cleanedLine;
+      return MARKUP_CHARS.reduce((acc, val) => {
+        return acc.replaceAll(`\u{FFFE}${val.codePointAt(0)}\u{FFFF}`, val);
+      }, cleanedLine);
     }
   }
 
diff --git a/website/_dogfooding/_docs tests/doc-with-complex-description.mdx b/website/_dogfooding/_docs tests/doc-with-complex-description.mdx
new file mode 100644
index 000000000000..ebb5e46e85c6
--- /dev/null
+++ b/website/_dogfooding/_docs tests/doc-with-complex-description.mdx	
@@ -0,0 +1,3 @@
+# Doc with Complex Description
+
+One <strong>Two</strong> _Three_ `<Four>` `:Five:` <em>Six</em> ![Seven](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAgAAAAIAQMAAAD+wSzIAAAABlBMVEX///+/v7+jQ3Y5AAAADklEQVQI12P4AIX8EAgALgAD/aNpbtEAAAAASUVORK5CYII)
diff --git a/website/_dogfooding/docs-tests-sidebars.js b/website/_dogfooding/docs-tests-sidebars.js
index d8dac0c10e50..da20f4e6a9b7 100644
--- a/website/_dogfooding/docs-tests-sidebars.js
+++ b/website/_dogfooding/docs-tests-sidebars.js
@@ -22,6 +22,7 @@ const sidebars = {
     },
     'doc-without-sidebar',
     'doc-with-another-sidebar',
+    'doc-with-complex-description',
     'doc-with-last-update',
     {
       type: 'category',