1. **Prompt Fuzzer**: Interactive tool to evaluate and fortify system prompts against dynamic attacks, with a Playground chat interface for refining prompts iteratively.
2. **Garak**: Tests for vulnerabilities in LLMs, including hallucination, data leakage, and misinformation, continuously enhanced to better support applications.
3. **HouYi**: Automatically injects prompts into LLM-integrated applications to test their vulnerability, including a demo script for simulation.
4. **JailbreakingLLMs**: Automatically generates jailbreak prompts for LLMs without human help, demonstrating high success rates and effectiveness.
5. **LLMAttacks**: Effectively prompts models to produce undesirable outputs, highlighting significant vulnerabilities in LLMs.
6. **PromptInject**: Framework for creating adversarial prompts for GPT-3, focusing on goal hijacking and prompt leaking.
7. **LLM Canary**: Open-source security benchmarking suite to identify and evaluate potential vulnerabilities in LLMs, incorporating OWASP Top 10 for LLMs.
8. **PyRIT**: Library designed to enhance the robustness evaluation of LLM endpoints, automating AI red teaming tasks and identifying security and privacy harms.
9. **LLMFuzzer**: Open-source fuzzing framework tailored for LLMs and their API integrations, ideal for uncovering and exploiting vulnerabilities in AI systems.
10. **PromptMap**: Automates the testing of prompt injection attacks by analyzing the context and purpose of ChatGPT rules, helping identify and mitigate potential vulnerabilities by simulating real attack scenarios.
11. **Gitleaks**: Static Application Security Testing (SAST) tool designed to detect hardcoded secrets in git repositories.
12. **Cloud_enum**: Multi-cloud OSINT tool designed to identify public resources across AWS, Azure, and Google Cloud, offering comprehensive enumeration of resources in each platform.
For more information and free consultation, you can reach out to our AI Lab in Telegram @aiscrumbot or follow us on Twitter @itinaicom.
