Efficient traffic management is essential for urban mobility, and real-time vehicle monitoring plays a vital role in reducing congestion and enhancing road safety. There have been various approaches based on vehicle detection and tracking models that calculate metrics like traffic density and vehicle counts in order to create a traffic monitoring system. Separately, the problem of Traffic Signal Control (TSC) has also been studied extensively with recent reinforcement learning (RL) based approaches showing promising results. In this paper, we propose TrafficMan, a framework to integrate existing vehicle tracking based traffic monitoring systems with reinforcement learning based TSC methods to create a unified system for traffic management. The predictive agent is first trained using RL algorithms for TSC in a simulation environment. A method for integrating existing systems to the environment for utilizing the RL agent is devised. Finally, the system is evaluated for different combinations of vehicle detection and tracking models and RL models to measure performance. The framework aims to extend traffic management systems by integrating RL agents for TSC thereby providing more precise and responsive control over traffic dynamics.