This project is a sample chatbot powered by a Large Language Model (LLM) ReAct agent, implemented with Langchain. It’s designed to be an educational tool for security researchers, developers, and enthusiasts to understand and experiment with prompt injection attacks in ReAct agents.

The project specifically focuses on Thought/Action/Observation injection, as described in the WithSecure Labs publication and accompanying video tutorial.

This repository is an adaptation of a challenge created by WithSecure for the Capture The Flag (CTF) competition held at BSides London 2023.