Multiple page loads
Snapshots are captured after independent reloads, so an accidental DOM match is not treated as stable.
Chrome extension · DOM analysis
Reliable selectors for changing pages
StableHTMLFinder compares page structure across independent loads and reveals which elements can be trusted in automation, testing, and filtering rules.
Discuss a similar projectThe challenge
Why a selector can fail after a reload
Many websites use obfuscated HTML, dynamically generated classes and IDs, changing nesting, and disposable utility elements. These techniques can also make advertising blocks harder to detect and block. A CSS selector created from a single page load often includes transient values and stops finding the element after a reload. StableHTMLFinder displays the page structure together with element characteristics that persist across multiple loads.
Stability map
The map reveals persistent tags, attribute names and values, classes, text, and other structural characteristics.
Complex boundaries
Accessible iframe and Shadow DOM content remains part of the analysis, together with the context in which an element was found.
CSS and BlockIt rules
The extension can suggest a standard CSS selector or a BlockIt rule. BlockIt extends selector syntax with characteristics CSS cannot express: attribute and class counts, own text, and patterns inside attribute names or values.
What makes it different
Selectors based on evidence from multiple page loads
Most selector tools inspect an element only in the current DOM. That may be enough for a conventional page, but with obfuscated HTML the resulting selector often depends on a transient class, ID, or position. StableHTMLFinder first reveals which characteristics actually persist across page loads, then lets the user build a rule around them. Useful characteristics are not limited to what standard CSS selectors can describe.
Built for
A tool for the web as it really is
- BlockIt users who want to hide elements on a website
- QA engineers and UI test authors
- Browser extension developers
- Web automation specialists
- Filter and blocking-rule authors
How it works
From a live page to a verified landmark
Capture
The extension collects the page structure several times with a configurable delay.
Analyze
The extension identifies which parts of the structure persist and which change from one load to the next.
Map
The result becomes an interactive tree of stable and changing regions.
Select
Find the target element and create a selector based on verified structure.
Technology
A lightweight, framework-free architecture
Chrome Extension Manifest V3JavaScript ES ModulesChrome APIsDOM snapshot analysisiframe & Shadow DOMNode.js build scriptsNode.js + jsdom testsLocalization in 8 languages