aivillage / llm_verification

LLM prompt attacks for hacker CTFs via CTFd.
https://aivillage.org
8 stars 1 forks source link

Proposed Roadmap #16

Closed canyon289 closed 10 months ago

canyon289 commented 1 year ago

Jotting down ideas, feel free to edit directly or comment

Development

Testing

Simulation of faulty network - Potentially use https://hub.docker.com/r/chaostoolkit/chaostoolkit

comath commented 1 year ago

Started working on the multiturn: https://github.com/aivillage/llm_verification/pull/17