This paper introduces a software framework developed for analyzing EEG signal using machine learning methods. The framework consists of several independent and customizable modules for signal acquisition and preprocessing, feature extraction, model training, evaluation, and interpretation. A unique aspect is the flexibility to tune hyperparameters across all stages of preprocessing and feature extraction. The framework was applied to two tasks: diagnosis of mental disorders and detection of intention to perform a hand movement. The results demonstrate balanced accuracy rates of 91% for schizophrenia diagnosis, 88% for obsessive-compulsive disorder diagnosis and 77% for movement intention detection. The methodologies employed for both tasks are detailed in the study. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.