RIMA (Rapping in Italy: Music and Artists) is a structured dataset that includes information and features related to the songs of 104 Italian rap artists, both male and female. Designed in a database-like format, it supports research in fields such as natural language processing, music information retrieval, sociolinguistics, and digital humanities. Each artist in the dataset is assigned a unique identifier, stored in a central indexing file that ensures consistent cross-referencing across all components. A separate metadata file provides demographic and geographic details about the artists, such as gender, birth year, years active, and place of origin. These data are primarily sourced from Wikipedia. For each artist, the dataset contains their full discography, where each song is represented by 199 features. These features are organized into multiple categories. Metadata fields include internal identifiers as well as external IDs from platforms like Genius and Spotify. The lyrics of each song are included in full, accompanied by a rich set of linguistic features mainly extracted using the Profiling-UD framework. These cover syntactic and morphological properties, lexical statistics, and explicit information on the presence and list of profanities in both Italian and English. In addition, audio-derived features are provided, describing aspects such as tempo, signal amplitude, rhythmic patterns, and energy dynamics. By integrating textual, linguistic, and acoustic data, RIMA enables a wide range of analyses, including artist profiling, stylistic and sociolinguistic studies, gender-based comparisons, and multimodal modeling of musical and lyrical content within the Italian rap genre.