Our papers on jailbreaking LLMs and agent-based secure code generation presented in workshops @ NeurIPS 2024