Work in Progress
This is a browser extension for removing hiragana hints from the Japan Times Language articles.
Right now, the script is hard-coded to Jōyō [常用] kanji levels, since that was the easiest data for me to grab and test against.
The script works by looking for a sequence of kanji followed by hiragana in parentheses, e.g., "困(こま)る". It checks to see if all of the kanji are known kanji, and if so, removes the parenthetical hiragana. If one kanji is known but another is not in a compound word, it will retain the hiragana hint. Additionally, it is only removing pure-hiragana hints. This prevents removing parenthetical statements which include meanings the article is purposefully trying to convey, e.g., "伝える(to tell)" or "全員(ぜんいん, everyone, all members)".
I find the language articles at the Japan Times website to be useful, but since I already know quite a bit of kanji, I find the hiragana hints distracting when they are for kanji I already know. Furthermore, having the hint undermines a chance to test my learning, i.e., that I DO know how to read the kanji without the hint, since I cannot unsee the hint.
My goal is to have this browser extension be dynamic, potentially even sharing it with others if they find it useful. This will mean allowing the user to set their kanji reading level. I may also try to tie it into WaniKani levels, so that it queries a user's API key then eliminates kanji based upon kanji learned up through the user's current level.