HanGrid
HanGrid is an incremental puzzle game about building Chinese characters from radicals.
Buy components, place them on a 5x5 grid, and arrange them so nearby parts form valid Chinese characters. Each component generates income, but completed characters become stronger engines for growth.
Examples:
亻 + 木 = 休
氵 + 古 + 月 = 湖
The game is still an early prototype, focused on testing whether Chinese character composition can work as an incremental game loop.
| Status | In development |
| Platforms | HTML5 |
| Rating | Rated 2.0 out of 5 stars (3 total ratings) |
| Author | chunq |
| Genre | Puzzle, Educational |
| Tags | chinese, Idle, Incremental, Word game |
| Links | Steam |
| Content | No generative AI was used |
Development log
- HanGrid 0.4.1 Update: Free Shop Components37 days ago
- HanGrid 0.3.0 is live.43 days ago
- HanGrid 0.2.0: Tutorial, voice support, and clearer combination hints43 days ago

Comments
Log in with itch.io to leave a comment.
If you were to sell your characters because they were making to little for example, and you don't have enough money to buy another your just stuck and have to restart, I hadn't realized that and sold all mine
Yeah, I should probably make that much clearer.
Deleting components is mainly meant to free up board space for better combinations, not as a way to recover money. Right now it does not refund any currency, so selling everything can softlock the run if you can’t afford another component.
I’ll add a warning/tutorial hint for that.
a way to handle it might be to have the cost based off number of slots filled rather than just increasing based on each purchase. being able to move purchased characters in the board would also be nice
That makes sense, thanks for the suggestion.
Basing the price on filled board slots instead of total purchases would probably make the system feel less punishing, especially after deleting components. I’ll look into changing the pricing logic that way.
Moving purchased components on the board is also a good point. Right now placement is too final, so I agree that some kind of move/reposition feature would make experimentation much better.