aivillage/llm_verification

LLM prompt attacks for hacker CTFs via CTFd.

19
/ 100
Experimental

This plugin helps security event organizers create challenges for participants to test and "attack" Large Language Models (LLMs). Organizers can define challenges where participants craft prompts to make an LLM generate specific, often 'cheeky' or subverting, responses. The organizers then manually review these submitted prompts and their LLM outputs, assigning points based on how effectively the participant's prompt subverted the model.

No commits in the last 6 months.

Use this if you are organizing a Capture The Flag (CTF) or similar cybersecurity competition and want to include challenges focused on red-teaming or subverting AI language models.

Not ideal if you are looking for an automated system to evaluate LLM outputs without human intervention, or if your competition does not use the CTFd platform.

cybersecurity-competition CTF LLM-red-teaming AI-security prompt-engineering
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 5 / 25

How are scores calculated?

Stars

15

Forks

1

Language

Python

License

Last pushed

Dec 17, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/aivillage/llm_verification"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.