The search for alternative teaching-learning processes that attract more interest and involvement of young people, has inspired the development of a game with a chatbot architecture based on interactive storytelling and multiple learning paths. Thus, we introduce in this article the GameBot ZoAm, developed for the Discord instant messaging and social platform. ZoAm offers a unique learning experience centered around storytelling, focusing on fundamental computing concepts and logical challenges that enhance computational thinking skills. Furthermore, the game also promotes an appreciation for Amazonian culture and folklore, with decision-making with human values. An action research study was conducted involving students from the last years of the end of elementary school. The research utilized a heuristic analysis based on the Gameplay Heuristics (PLAY) by Desurvire and Wiberg (ANO), and the evaluation model proposed by Korhonen and Koivisto (ANO) for mobile devices. The analysis employed a reduced and merged set of heuristics from these models, suited for the gamebot’s context, focusing on I) Usability, II) Gameplay and Immersion, and III) Mobility. Regarding the reliability coefficient used to evaluate the survey applied to students after playing the gamebot, Cronbach’s Alpha and Guttman Lambda-6 (G6(smc)) coefficients were applied. These metrics were chosen to ensure the internal consistency and reliability of survey items, reflecting on how effectively the questions measured the focuses proposed by the heuristic analysis. The findings indicate that the game has the potential to facilitate the assimilation of the integrated concepts and sustain student interest throughout gameplay.