We present a new technique that enables people to create shape-bounded Wordle, fitting text into a given shape. To guide the word placement within a shape, we devise shape-aware Archimedean spirals by extending traditional Archimedean spirals with the distance field of the shape. To handle a non-convex shape, we introduce a multi-centers Wordle layout by segmenting the shape into multiple regions and then applying our shape-aware spirals to generate Wordle for each region. To efficiently fill the shape while allowing the editing of important words, we use the bounding boxes for important words and fill up the white space with the pixel images of less important ones. We also offer a set of editing interactions to facilitate the creation of semantically meaningful Wordle. In addition, we present three forms of evaluation: comprehensive quantitative comparisons of our results against those of the state-of-the-art technique (WordArt), case studies with designers, and a gallery to demonstrate our technique's coverage.