# Awesome LLMSecOps

A curated list of resources for LLM security operations, covering architecture risks, vulnerabilities, jailbreaks, RAG security, defensive tools, threat modeling, datasets and research.

Awesome LLMSecOps organizes resources under headings including architecture risks, vulnerabilities, tools, defense, threat modeling, jailbreaks, RAG security, proofs of concept, study resources, frameworks, best practices, research and tutorials, aimed at building an LLM development process that accounts for its security risks.

It documents architectural risk categories such as recursive pollution from feeding model output back into training data, data debt from unvetted training datasets, black-box opacity in foundation models, prompt manipulation, and data poisoning, alongside links to tools and further reading for each. The maintainers note that, unlike similar lists, it emphasizes practical security implementation over links to research papers.

It is presented as a curated list rather than runnable software.

Website: https://github.com/wearetyomsmnv/Awesome-LLMSecOps