diff --git a/src/language/markdown-source-code.js b/src/language/markdown-source-code.js
index 2c45f96b..ddb8f84c 100644
--- a/src/language/markdown-source-code.js
+++ b/src/language/markdown-source-code.js
@@ -13,6 +13,7 @@ import {
 	ConfigCommentParser,
 	Directive,
 } from "@eslint/plugin-kit";
+import { lineEndingPattern } from "../util.js";
 
 //-----------------------------------------------------------------------------
 // Types
@@ -152,7 +153,7 @@ export class MarkdownSourceCode extends TextSourceCodeBase {
 	 * @param {Root} options.ast The root AST node.
 	 */
 	constructor({ text, ast }) {
-		super({ ast, text });
+		super({ ast, text, lineEndingPattern });
 		this.ast = ast;
 
 		// need to traverse the source code to get the inline config nodes
diff --git a/src/util.js b/src/util.js
index 36cf9954..b69e80eb 100644
--- a/src/util.js
+++ b/src/util.js
@@ -7,6 +7,12 @@
 // Regex Patterns
 //-----------------------------------------------------------------------------
 
+/**
+ * Line ending pattern to match all line endings (CRLF, CR, LF). (CommonMark spec)
+ * @see https://spec.commonmark.org/0.31.2/#line-ending
+ */
+export const lineEndingPattern = /\r\n|[\r\n]/u;
+
 /**
  * CommonMark does not allow any white space between the brackets in a reference link.
  * If that pattern is detected, then it's treated as text and not as a link. This pattern
@@ -24,16 +30,16 @@ export const htmlCommentPattern = /<!--[\s\S]*?-->/gu;
 //-----------------------------------------------------------------------------
 
 /**
- * Checks if a frontmatter block contains a title matching the given pattern
- * @param {string} value The frontmatter content
- * @param {RegExp|null} pattern The pattern to match against
- * @returns {boolean} Whether a title was found
+ * Checks if a frontmatter block contains a title matching the given pattern.
+ * @param {string} value The frontmatter content.
+ * @param {RegExp|null} pattern The pattern to match against.
+ * @returns {boolean} Whether a title was found.
  */
 export function frontmatterHasTitle(value, pattern) {
 	if (!pattern) {
 		return false;
 	}
-	const lines = value.split("\n");
+	const lines = value.split(lineEndingPattern);
 	for (const line of lines) {
 		if (pattern.test(line)) {
 			return true;
diff --git a/tests/language/markdown-source-code.test.js b/tests/language/markdown-source-code.test.js
index b4011af7..daac0ca2 100644
--- a/tests/language/markdown-source-code.test.js
+++ b/tests/language/markdown-source-code.test.js
@@ -66,6 +66,67 @@ describe("MarkdownSourceCode", () => {
 		sourceCode = new MarkdownSourceCode({ text: markdownText, ast });
 	});
 
+	describe("constructor", () => {
+		it("should create a MarkdownSourceCode instance", () => {
+			assert.strictEqual(
+				sourceCode.constructor.name,
+				"MarkdownSourceCode",
+			);
+			assert.strictEqual(sourceCode.ast, ast);
+			assert.strictEqual(sourceCode.text, markdownText);
+		});
+	});
+
+	describe("lines", () => {
+		it("should parse CRLF line endings", () => {
+			const text = "lumir\r\nlumir";
+			const sourceCodeWithCRLF = new MarkdownSourceCode({
+				text,
+				ast: fromMarkdown(text),
+			});
+
+			assert.deepStrictEqual(sourceCodeWithCRLF.lines, [
+				"lumir",
+				"lumir",
+			]);
+		});
+
+		it("should parse CR line endings", () => {
+			const text = "lumir\rlumir";
+			const sourceCodeWithCR = new MarkdownSourceCode({
+				text,
+				ast: fromMarkdown(text),
+			});
+
+			assert.deepStrictEqual(sourceCodeWithCR.lines, ["lumir", "lumir"]);
+		});
+
+		it("should parse LF line endings", () => {
+			const text = "lumir\nlumir";
+			const sourceCodeWithLF = new MarkdownSourceCode({
+				text,
+				ast: fromMarkdown(text),
+			});
+
+			assert.deepStrictEqual(sourceCodeWithLF.lines, ["lumir", "lumir"]);
+		});
+
+		it("should parse CRLF CR LF line endings", () => {
+			const text = "lumir\r\nlumir\rlumir\nlumir";
+			const sourceCodeWithCRLFCRLF = new MarkdownSourceCode({
+				text,
+				ast: fromMarkdown(text),
+			});
+
+			assert.deepStrictEqual(sourceCodeWithCRLFCRLF.lines, [
+				"lumir",
+				"lumir",
+				"lumir",
+				"lumir",
+			]);
+		});
+	});
+
 	describe("getText()", () => {
 		it("should return the text of the Markdown source code", () => {
 			assert.strictEqual(sourceCode.getText(), markdownText);
diff --git a/tests/util.test.js b/tests/util.test.js
index f5c2de57..ccaaf5f8 100644
--- a/tests/util.test.js
+++ b/tests/util.test.js
@@ -36,7 +36,25 @@ describe("util", () => {
 			);
 		});
 
-		it("should return true if the pattern matches any line in multiline frontmatter", () => {
+		it("should return true if the pattern matches any line in multiline frontmatter (CRLF)", () => {
+			const frontmatter = [
+				"description: Test",
+				"title: My Document",
+				"author: lumirlumir",
+			].join("\r\n");
+			assert.strictEqual(frontmatterHasTitle(frontmatter, pattern), true);
+		});
+
+		it("should return true if the pattern matches any line in multiline frontmatter (CR)", () => {
+			const frontmatter = [
+				"description: Test",
+				"title: My Document",
+				"author: lumirlumir",
+			].join("\r");
+			assert.strictEqual(frontmatterHasTitle(frontmatter, pattern), true);
+		});
+
+		it("should return true if the pattern matches any line in multiline frontmatter (LF)", () => {
 			const frontmatter = [
 				"description: Test",
 				"title: My Document",