EvEMTBench is a research benchmark (synthetic power-grid EMT data + evaluation code); it processes no secrets and has no network/authentication surface. Still, we take defects seriously.

Please report suspected security issues (e.g. a crafted input file that triggers unsafe file writes, code execution during data loading, or a dependency CVE that affects this package) privately to:

• Julian Oelhaf — julian.oelhaf@fau.de

Do not open a public issue for a security report. Include a description, affected version (python -c "import evemtbench; print(evemtbench.__version__)" / BENCHMARK_VERSION), and a minimal reproduction. We aim to acknowledge within 5 working days.

The main branch and the latest tagged BENCHMARK_VERSION are supported. Older frozen benchmark versions are retained read-only for reproducibility and do not receive fixes.