Modern linguistic tools for Ukrainian, such as GRAC [7], rely on extensive lexical resources like VESUM [5], the largest online dictionary containing over 418,000 lemmas. However, these tools face notable limitations when encountering forms derived from autonomous lexical units (ALUs) (We adopt the term “ALU” (Autonomous Lexical Unit) as proposed by Max Silberztein [8]).) that are not included in existing dictionaries, requiring constant updates to maintain coverage. Moreover, current systems treat derived forms and their base forms as separate entries, without establishing explicit links between them. Ukrainian poses particular challenges in this respect, especially with derived adjectives. Relative adjectives can be formed from proper names, such as place names or personal names. While major toponyms (e.g., kyjivsʹkyj (Ukrainian is written in the Cyrillic alphabet, however, for the sake of readability, we will use the traditional scholarly transliteration for Ukrainian examples and references. When it is necessary to provide the word in Cyrillic to explain an image, we will include the transliteration alongside it, as in нськ /ns'k/.) [from Kyiv]) are covered, derivatives from smaller towns or personal names are usually omitted. Other productive adjective types—diminutive, excessive, and short forms—are also underrepresented, even though they appear frequently in written language. Participles present an additional difficulty due to their hybrid verbal–adjectival nature. Existing tools fail to link participles to their source verbs, resulting in incomplete morphological and semantic analysis. To address these limitations, we have developed several morphological grammars that accurately annotate derived forms—including relative, diminutive, excessive, and short adjectives, as well as participles. These grammars systematically connect derived forms to their base forms, improving both dictionary completeness and linguistic tool performance. This work strengthens Ukrainian NLP resources and contributes to more reliable applications for linguistic research and automatic text processing.