The use of personality to improve recommendations is a growing trend in recommender systems. However, to accurately determine someone's personality is complex, such as the need for long personality questionnaires which are subject to social desirability bias, or the need for a great amount of users' interactions with the system. Also, most of the existing works focus on obtaining the broader personality dimensions instead of the more granular traits, which better characterize a person, and, so far, there are no shortduration mobile games that can accurately predict personality. In this work, we propose to implicitly acquire the users' more granular personality traits, namely cautiousness and achievementstriving as a first concept proof, by using mobile short-duration serious games, in an attempt to also replace self-reporting personality questionnaires. A simulation with real participants (𝑛=100) was conducted. Several significant relationships with the proposed traits were found, although the sample size and social desirability bias found in the self-reporting personality questionnaire seemed to hinder the obtention of significant correlations. Interesting significant correlations with anger, modesty, friendliness, excitement seeking, cheerfulness, and adventurousness personality traits were also found. The results show mobile minigames are a viable way of unobtrusively determining the  users’ more granular personality, being the path to replace personality questionnaires.