modelcontextprotocol
diff --git a/‎src/fetch/src/mcp_server_fetch/server.py‎
Lines changed: 9 additions & 1 deletion b/‎src/fetch/src/mcp_server_fetch/server.py‎
Lines changed: 9 additions & 1 deletion
diff --git a/‎src/fetch/tests/test_server.py‎
Lines changed: 37 additions & 0 deletions b/‎src/fetch/tests/test_server.py‎
Lines changed: 37 additions & 0 deletions
@@ -41,7 +41,15 @@ def extract_content_from_html(html: str) -> str:
     content = markdownify.markdownify(
         ret["content"],
         heading_style=markdownify.ATX,
-    )
+    ).lstrip()
+    title = ret.get("title")
+    if title:
+        title_markdown = f"# {title.strip()}"
+        # Readability often omits the document title when the article body does
+        # not repeat it. Include it once so callers can identify fetched pages,
+        # while avoiding duplicate headings when the title is already present.
+        if title_markdown.casefold() not in content[: len(title_markdown) + 32].casefold():
+            content = f"{title_markdown}\n\n{content}"
     return content
 
 
 
@@ -67,6 +67,43 @@ def test_simple_html(self):
         # readabilipy may extract different parts depending on the content
         assert "test paragraph" in result
 
+    def test_html_includes_document_title_when_omitted_from_article(self):
+        """Test that simplified markdown preserves the page title."""
+        html = """
+        <html>
+        <head><title>What’s new in 2.1.0 (Aug 30, 2023)</title></head>
+        <body>
+            <article>
+                <p>These are the release notes.</p>
+            </article>
+        </body>
+        </html>
+        """
+
+        result = extract_content_from_html(html)
+
+        assert result.startswith("# What’s new in 2.1.0 (Aug 30, 2023)")
+        assert "These are the release notes." in result
+
+    def test_html_does_not_duplicate_existing_title_heading(self):
+        """Test that a page title already present as the first heading is not duplicated."""
+        html = """
+        <html>
+        <head><title>Existing Title</title></head>
+        <body>
+            <article>
+                <h1>Existing Title</h1>
+                <p>Article body.</p>
+            </article>
+        </body>
+        </html>
+        """
+
+        result = extract_content_from_html(html)
+
+        assert result.count("# Existing Title") == 1
+        assert "Article body." in result
+
     def test_html_with_links(self):
         """Test that links are converted to markdown."""
         html = """