gdalmau/lakera-gandalf-solutions

My inputs for the LLM Gandalf made by Lakera

28
/ 100
Experimental

This project provides example prompts to bypass the safety mechanisms of the 'Gandalf LLM' game. It offers a collection of inputs designed to extract a secret password from the AI, despite its programming to withhold it. Anyone interested in understanding AI limitations and prompt engineering for security testing would find this useful.

No commits in the last 6 months.

Use this if you are playing the Gandalf LLM game and need strategies to extract the hidden passwords from each level.

Not ideal if you are looking for general-purpose prompt engineering techniques or a tool for daily AI interaction, as it's specific to one game.

AI game challenge prompt engineering examples AI security testing LLM interaction strategies AI safety bypass
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 12 / 25

How are scores calculated?

Stars

48

Forks

6

Language

License

Last pushed

Sep 18, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/gdalmau/lakera-gandalf-solutions"

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