Acknowledgments¶

This repository draws from multiple sources, including academic research, community efforts, and existing projects that document prompt injection techniques. We would like to acknowledge the following:

Projects¶

Garak: A tool for LLM vulnerability scanning and penetration testing
Denzel-Crocker-Hunting-For-Fairly-Odd-Prompts: Collection of jailbreak templates
PromptInject: Framework for testing LLM prompts against adversarial inputs

Research¶

We’ve incorporated insights from various research papers and publications in the field of LLM security:

Various academic papers on prompt injection techniques and defenses
Industry security reports documenting LLM vulnerabilities
Security advisories from AI system providers

Community Contributors¶

This repository has benefited from the contributions and feedback of many individuals in the AI security community. We appreciate everyone who has contributed examples, suggested improvements, or helped classify and document techniques.

Citing This Work¶

If you use this resource in your research or projects, please cite:

@misc{prompt-injections,
  author = {Astley, William},
  title = {Prompt Injection Techniques Repository},
  year = {2025},
  publisher = {GitHub},
  howpublished = {\url{https://github.com/pr1m8/prompt-injections}}
}