The emergence of Industry 5.0 emphasizes human-centric manufacturing and increases the demand for product personalization, which introduces additional complexity to manual assembly processes. These challenges often require operators to memorize intricate sequences for various products, increasing the likelihood of human error and affecting production quality. In response to this, we present a modular framework developed in Python that leverages computer vision algorithms to monitor each step of the manual assembly process in real-time. The system features a low-code interface, allowing users to digitize and configure workflows without deep technical knowledge. Unlike hardware-dependent solutions, our approach offers a flexible, cost-effective alternative that integrates with existing production environments. A preliminary case study conducted in a controlled setting demonstrated the framework’s ability to detect errors during specific assembly tasks. The results indicate its potential to enhance quality control and support operator performance. Future work will focus on applying the framework to complete the assembly process and incorporating augmented reality to provide detailed, step-by-step operator guidance, further reducing errors and improving process efficiency.