Goochbeater/Spiritual-Spell-Red-Teaming

A repo for jailbreaking various LLMs, mainly Claude

What it solves

This project addresses the need for security testing (red teaming) of AI models, specifically focusing on identifying vulnerabilities that could allow users to bypass safety filters and generate restricted content.

How it works

The project provides a collection of techniques and prompts designed to test the robustness of AI models. It focuses on "spiritual" or creative red teaming, using complex prompt engineering to discover how models can be manipulated into ignoring their safety guidelines.

Who it’s for

It is intended for security researchers and AI developers who want to evaluate the safety and alignment of their models by simulating adversarial attacks.

Highlights

  • Focuses on red teaming and security research for AI models.
  • Provides methods to test safety filter bypasses.
  • Aims to improve model robustness through adversarial testing.

Related

  • Project
    confident-ai/deepteam

    An open-source red teaming framework for LLM systems that simulates adversarial attacks to uncover vulnerabilities and provides guardrails for production security.

  • Project
    AISecurityLab/hackagent

    A security red-teaming toolkit for AI agents that automates the discovery of vulnerabilities like prompt injection and jailbreaking using research-backed attack techniques.

  • Project
    Tencent/AI-Infra-Guard

    An AI red teaming platform that provides automated security scanning for AI infrastructure, agent skills, and LLM jailbreak evaluation to identify vulnerabilities and risks.

  • Dispatch
    Anthropic outlines challenges and best practices for red teaming AI systems

    Anthropic released a detailed analysis of red‑team­ing methods—expert, automated, multimodal, and crowdsourced—highlighting their benefits, challenges, and policy recommendations to standardise AI safety testing.

  • Dispatch
    OpenAI Red Teaming Network

    OpenAI has launched the Red Teaming Network, a formal community of global domain experts tasked with rigorously evaluating AI models to identify risks and improve safety throughout the development lifecycle.