Identification of JavaScript files with Public Class Fields ...

Dear GitHub support,

could you please make JavaScript files with Public Class Fields, Private Elements and Static Initialization Blocks identifiable again?

The irreplaceable but abandoned Waterfox Classic web browser (which is based on Firefox 56) does not support these web features. As a workaround, the unofficial [Polly+ extension](https://github.com/BrowserWorks/Waterfox-Classic/issues/182#issuecomment-1362129403) can transform JavaScript files on loading using the [Babel standalone](https://babeljs.io/docs/babel-standalone) script to avoid errors. This has been the case since late 2023, [after the compatible service was removed from GitHub](https://github.com/martok/palefill/issues/94#issuecomment-1854367897).

Previously, only URLs containing specific substrings were transformed (such as ``modules_oddbird_popover|modules_github_arianotify|modules_github_text-expander-element|modules_tanstack_|modules_date-fns_addWeeks|/repositories-|modules_emotion_is|_memoize_dist|_hotkey_dist|_primer_live-region-e|_hydro-analyt|_accname_dist|/environment-|e-attachment-e|/commits-|combobox-nav_|_lit-labs|_g-emoji-e|_quote-selection_|_mergebox_i|-alive-subscription_|_handlewithcare_r|copilot-|/issues-react-|-relay_hooks|dom-input-range_|emoji-regex_|focus-visible_|swc_helpers``), where a script would otherwise have caused errors.

In November 2025, the meaningful JavaScript filenames were replaced with random characters (example: https://github.githubassets.com/assets/2785-addb5ec08646.js).

Please tell me how I can identify files that use the aforementioned web functions for performance reasons.

Thanks.

The issue names no repository files, tests, or entry points. Start by determining whether browser-support contains the detection or transformation logic, then define a reproducible check for JavaScript files using Public Class Fields, Private Elements, or Static Initialization Blocks; done requires an agreed implementation scope and validation.

Written by the indexing model from the issue text.