kepano · mvanhorn · Jul 4, 2026 · Jul 4, 2026 · Jul 4, 2026
diff --git a/src/removals/selectors.ts b/src/removals/selectors.ts
@@ -9,13 +9,49 @@ import {
 	FOOTNOTE_LIST_SELECTORS
 } from '../constants';
 import { DebugRemoval } from '../types';
-import { textPreview, logDebug } from '../utils';
+import { textPreview, logDebug, normalizeText } from '../utils';
 import { getClassName, hasResponsiveShowClass } from '../utils/dom';
 
+const MAJORITY_CONTENT_MIN_TEXT_LENGTH = 200;
+const MAJORITY_CONTENT_RATIO = 0.8;
+const AMBIGUOUS_MAJORITY_EXACT_SELECTOR = [
+	'[class^="ad-" i]',
+	'[class$="-ad" i]',
+	'[id^="ad-" i]',
+	'[id$="-ad" i]'
+].join(',');
+const ALWAYS_REMOVABLE_EXACT_TAGS = new Set([
+	'SCRIPT',
+	'STYLE',
+	'NOSCRIPT',
+	'IFRAME',
+	'FRAME',
+	'FRAMESET',
+	'OBJECT',
+	'EMBED',
+	'APPLET',
+	'BASE',
+	'META',
+	'LINK'
+]);
+
+function isAlwaysRemovableExactElement(el: Element): boolean {
+	return ALWAYS_REMOVABLE_EXACT_TAGS.has(el.tagName.toUpperCase());
+}
+
+function isAmbiguousMajorityExactElement(el: Element): boolean {
+	return (
+		(el.tagName.includes('-') && el.matches('.ad:not([class*="gradient"])')) ||
+		el.matches(AMBIGUOUS_MAJORITY_EXACT_SELECTOR)
+	);
+}
+
 export function removeBySelector(doc: Document, debug: boolean, removeExact: boolean = true, removePartial: boolean = true, mainContent?: Element | null, debugRemovals?: DebugRemoval[], skipHiddenExactSelectors: boolean = false) {
 	const startTime = Date.now();
 	let exactSelectorCount = 0;
 	let partialSelectorCount = 0;
+	const mainContentTextLength = mainContent ? normalizeText(mainContent.textContent || '').length : 0;
+	const shouldGuardMajorityContent = mainContentTextLength >= MAJORITY_CONTENT_MIN_TEXT_LENGTH;
 
 	// Track all elements to be removed, with their match type
 	const elementsToRemove = new Map<Element, { type: 'exact' | 'partial'; selector?: string }>();
@@ -138,6 +174,16 @@ export function removeBySelector(doc: Document, debug: boolean, removeExact: boo
 		if (mainContent && el.contains(mainContent)) {
 			return;
 		}
+		if (
+			shouldGuardMajorityContent &&
+			type === 'exact' &&
+			!el.matches(HIDDEN_EXACT_SELECTOR) &&
+			!isAlwaysRemovableExactElement(el) &&
+			isAmbiguousMajorityExactElement(el) &&
+			normalizeText(el.textContent || '').length >= mainContentTextLength * MAJORITY_CONTENT_RATIO
+		) {
+			return;
+		}
 		if (el.tagName === 'A' && el.closest('h1, h2, h3, h4, h5, h6')) {
 			return;
 		}

diff --git a/tests/expected/issues--283-clutter-selector-removes-content.html b/tests/expected/issues--283-clutter-selector-removes-content.html
@@ -0,0 +1 @@
+<body>  <div>   <p> The Modular Supply Cart keeps shared equipment organized for small teams that need a predictable workflow without a dedicated storage room. </p> <H2>Description</H2> <p> Each cart ships with removable bins, a labeled front panel, and locking casters for stable positioning during inventory checks. The frame is designed for common shop, studio, and classroom supplies. </p> <H2>Features</H2> <ul> <li>Adjustable shelves with three height positions</li> <li>Four removable bins for cables, tools, and consumables</li> <li>Integrated label rail for team-specific organization</li> </ul> <H2>Specifications</H2> <ul> <li>Powder-coated steel frame with recycled plastic bins</li> <li>Supports up to 120 pounds of evenly distributed equipment</li> <li>Ships flat with hardware and assembly instructions</li> </ul> </div>   </body>
diff --git a/tests/expected/issues--283-clutter-selector-removes-content.md b/tests/expected/issues--283-clutter-selector-removes-content.md
@@ -0,0 +1,26 @@
+```json
+{
+  "title": "Modular Supply Cart",
+  "author": "",
+  "site": "Example Catalog",
+  "published": ""
+}
+```
+
+The Modular Supply Cart keeps shared equipment organized for small teams that need a predictable workflow without a dedicated storage room.
+
+## Description
+
+Each cart ships with removable bins, a labeled front panel, and locking casters for stable positioning during inventory checks. The frame is designed for common shop, studio, and classroom supplies.
+
+## Features
+
+- Adjustable shelves with three height positions
+- Four removable bins for cables, tools, and consumables
+- Integrated label rail for team-specific organization
+
+## Specifications
+
+- Powder-coated steel frame with recycled plastic bins
+- Supports up to 120 pounds of evenly distributed equipment
+- Ships flat with hardware and assembly instructions
diff --git a/tests/fixtures/issues--283-clutter-selector-removes-content.html b/tests/fixtures/issues--283-clutter-selector-removes-content.html
@@ -0,0 +1,60 @@
+<!-- {"url":"https://example.com/products/modular-supply-cart/"} -->
+<!doctype html>
+<html lang="en">
+<head>
+  <meta charset="utf-8" />
+  <meta property="og:site_name" content="Example Catalog">
+  <meta property="og:title" content="Modular Supply Cart">
+  <title>Modular Supply Cart | Example Catalog</title>
+</head>
+<body>
+  <header class="site-header">
+    <a href="/">Example Catalog</a>
+    <nav>
+      <a href="/products/">Products</a>
+      <a href="/support/">Support</a>
+    </nav>
+  </header>
+
+  <app-root class="ad" ng-version="17.0.0">
+    <script type="application/json">
+      {"payload":"Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML. Runtime state should not appear in extracted HTML."}
+    </script>
+    <h1>Modular Supply Cart</h1>
+    <p>
+      The Modular Supply Cart keeps shared equipment organized for small teams that need
+      a predictable workflow without a dedicated storage room.
+    </p>
+
+    <h2>Description</h2>
+    <p>
+      Each cart ships with removable bins, a labeled front panel, and locking casters for
+      stable positioning during inventory checks. The frame is designed for common shop,
+      studio, and classroom supplies.
+    </p>
+
+    <h2>Features</h2>
+    <ul>
+      <li>Adjustable shelves with three height positions</li>
+      <li>Four removable bins for cables, tools, and consumables</li>
+      <li>Integrated label rail for team-specific organization</li>
+    </ul>
+
+    <h2>Specifications</h2>
+    <ul>
+      <li>Powder-coated steel frame with recycled plastic bins</li>
+      <li>Supports up to 120 pounds of evenly distributed equipment</li>
+      <li>Ships flat with hardware and assembly instructions</li>
+    </ul>
+  </app-root>
+
+  <div class="print-copy" style="display: none">
+    <h1>Modular Supply Cart</h1>
+    <p>This duplicate print copy should be removed as hidden content.</p>
+  </div>
+
+  <footer>
+    <p>Example Catalog footer links</p>
+  </footer>
+</body>
+</html>
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		<body> <div> <p> The Modular Supply Cart keeps shared equipment organized for small teams that need a predictable workflow without a dedicated storage room. </p> <H2>Description</H2> <p> Each cart ships with removable bins, a labeled front panel, and locking casters for stable positioning during inventory checks. The frame is designed for common shop, studio, and classroom supplies. </p> <H2>Features</H2> <ul> <li>Adjustable shelves with three height positions</li> <li>Four removable bins for cables, tools, and consumables</li> <li>Integrated label rail for team-specific organization</li> </ul> <H2>Specifications</H2> <ul> <li>Powder-coated steel frame with recycled plastic bins</li> <li>Supports up to 120 pounds of evenly distributed equipment</li> <li>Ships flat with hardware and assembly instructions</li> </ul> </div> </body>