With the increasing use of software libraries and frameworks, optimizing energy consumption has become a critical challenge for developers and researchers. Energy profiling, particularly for specific modules, is essential for identifying inefficiencies and optimizing performance, but it remains a time-consuming and error-prone process. Traditional methods often involve running complex benchmarks and repeatedly monitoring energy consumption through manual scripting or function-level instrumentation, making the process tedious and disruptive to workflows.

To address this, we introduce \textit{Software Library Energy Meter (SLEM)}, an open-source tool built atop Intel’s Running Average Power Limit (RAPL) interface to measure energy consumption across software frameworks, libraries, and custom scripts written in Python. Unlike existing tools such as PyJoules, which require developers to manually annotate functions using decorators or wrappers, \textit{SLEM} automates this process entirely—it identifies, extracts, and profiles functions without modifying source code or re-compiling. Moreover, it provides a configurable batch execution framework with workload simulation, enabling realistic and reproducible energy measurements that existing RAPL-based tools lack.

We demonstrate \textit{SLEM}’s effectiveness through empirical experiments by replicating energy measurement studies from prior work and show a significant reduction in manual effort. The tool is particularly valuable in research environments where iterative testing and performance evaluation of energy experiments are required. A demonstration of the tool is available at https://youtu.be/oRsxfT_kNNk, and additional resources along with related artifacts can be accessed at https://rishalab.github.io/SLEM/.