This collection brings together a set of Arabic-focused models, datasets, and tools designed to improve the security and safety of LLMs