This research systematically evaluates how different prompting techniques influence the security quality of code generated by Large Language Models.

• Investigates the relationship between prompting strategies and secure code generation
• Evaluates multiple prompting techniques and their impact on vulnerability prevention
• Offers actionable insights for developers to improve security in AI-assisted programming
• Provides a framework for more effective collaboration with coding LLMs

For security teams, this research offers critical guidance on mitigating potential vulnerabilities when integrating AI-generated code into production systems, establishing best practices for secure LLM interactions.

Prompting Techniques for Secure Code Generation: A Systematic Investigation