This repo is for the GPT "LLM Security Oracle". The goal of this project is to create a community-curated list of leading industry research that can be queried through the LLM Security Oracle.
Feel free to create a PR if you would like research to be added or have suggestions for the prompt instructions. Prompt Engineering feedback is welcomed!
- Prompt Engineering of the GPT's instruction
- Curating the documents so that the oracle can analyze industry-leading research related to LLMs
- Document Optimizations - The more documents, the slower it responds. (good prompt engineering might be able to short-circuit some of this.
- Analyzing PDFs for indirect prompt injections