A few honest answers.
Is this a Chinese proficiency test?
It is a short reading and vocabulary check. The L1–L4 recommendations are provisional site bands, not validated proficiency levels. We do not estimate vocabulary size, HSK, CEFR, IQ or cognitive benefits.
What if I know very little Chinese?
Choose “Not sure” whenever you need to. A completed 0/10 is a real result. Your report may say “Not enough evidence yet.” You can play any puzzle tier without a test.
Can I continue later?
On this browser, yes, while the anonymous cookie lasts (up to 90 days). Current answers and game state are saved on the server. Full-board drafts stay in this browser. Clearing browser data or changing devices can lose access; we offer no account recovery.
Who sees my report or challenge score?
Your complete report is private. Joining a room separately allows anyone holding that invitation link to see your nickname and score summary. The public leaderboard needs a separate opt-in. Card exports contain only the summary you choose.
How do friend challenges work?
Finish a reading check for Quiz PK, or a puzzle for Family PK. Create a seven-day invitation for yourself and up to ten friends. Friends join at their own pace. Previous first results on exactly the same version and translation are reused, and ties share rank.
Is the leaderboard cheat-proof?
No. Anonymous sessions, limited questions and memorized answers make this an informal competition. Server scoring, rate limits, version locks and first-attempt rules reduce easy abuse, but a new browser can appear as a new player.
Can I share on WeChat or WhatsApp?
Save a card image, then copy the real challenge link and paste it into your chat app. If your browser supports system sharing, that option is available too. We cannot confirm that a share button resulted in a social post.
Does “Got it” prove I learned a word?
No. It is your self-rating at that moment. We do not claim lasting learning or provide a next-day memory test.
Where do the character notes come from?
The collection uses reviewed structural data from Make Me a Hanzi and dictionary cross-checks from CC-CEDICT, with project-authored practice sentences. Readings are context-dependent; individual characters can have more than one pronunciation.