“Skinsense” is a new development where we present a comprehensive approach that uses machine learning models to analyze real-time captured skin images or existing images and predict their cancerous or benign nature, while also includes features that indicate findings belonging to the severity category. Our method uses deep learning algorithms and powerful image processing to evaluate skin lesions in order to aid in the early identification and detection of skin cancer (Afza et al. in Sensors 22:799, 2022 [1]). “Skinsense” employs the usage of multiple deep learning techniques such as ResNet101, Baseline CNN, Inception-ResNet-v2, VGG-16, and InceptionV3 to compare the accuracy provided by each of these models (Aldhyani et al. in Diagnostics 12:2048, 2022 [2]). The ResNet101 model with added layers of convolutional computation is the primary model with highest accuracy (Mukadam and Patil in Skin cancer classification framework using enhanced super resolution generative adversarial network and custom convolutional neural network, 2023 [3]). We have also integrated a smart chatbot interface to enable users to interact after receiving the classification results. Users can search for treatments, prevention strategies, cures, etc., which can provide them with valuable information and guidance. By combining advanced technology with a user-friendly interface, we aim to create a complete tool that not only aids diagnosis but also educates and inspires people to maintain healthy skin (El-Khatib et al. in Deep learning-based methods for automatic diagnosis of skin lesions, 2020 [4]).