<p>Current universal screening tools mainly assess reading and neglect spelling, particularly lexical spelling. This study examined the predictive performance of a multi-component screening protocol for the early identification of spelling difficulties in kindergarten, within a response-to-intervention (RTI) framework. The protocol, including tasks assessing first name writing, visual attention, letter knowledge, phonological awareness, and orthographic lexical memory (OLM), was administered at three time points to 129 kindergarten children. Lexical and phonological spelling were assessed in first grade (<i>N</i> = 98). ROC curve and logistic regression analyses were conducted to determine the protocol’s discriminatory capacity, sensitivity, and specificity. The protocol showed acceptable discriminatory capacity for lexical spelling at mid- and end-of-year screening points (AUC = 0.74–0.76), whereas performance at the beginning of kindergarten was more limited (AUC = 0.68). Letter knowledge was the most robust predictor across all time points and both spelling outcomes. OLM delayed recall was retained as a predictor of lexical spelling risk at mid-kindergarten, alongside letter knowledge, suggesting that orthographic memory may contribute to early identification of lexical spelling difficulties in opaque orthographies. These findings provide preliminary support for this protocol as an early screening tool that explicitly incorporates lexical spelling. The contribution of OLM appears modest and specific to lexical spelling. The protocol is compatible with RTI principles and may be relevant for screening in opaque orthographies.</p>