Projects sharing features with PyRIT

Giskard is an evaluation framework, testing library, and quality monitoring system for large language models and AI agents. It serves as a toolkit for quantifying model performance and reliability, providing specialized capabilities for validating retrieval-augmented generation pipelines. The project distinguishes itself through an automated red teaming tool and security scanner designed to identify vulnerabilities, prompt injections, and safety risks. It utilizes adversarial probing and synthetic edge case generation to quantify model robustness and detect information disclosure. The platfo

AI-Infra-Guard is a security scanning platform designed to detect vulnerabilities across large language model deployments, AI agent skills, and the underlying infrastructure. It functions as a security toolset for auditing source code, evaluating model robustness, and identifying insecure network configurations. The project provides a red teaming framework that uses curated attack datasets to test for jailbreak vulnerabilities and prompt injections. It also includes an infrastructure auditor that employs network fingerprinting and asset discovery to match running components against known comm

PurpleLlama is a collection of security components and toolkits designed for large language models. It provides specialized systems including a code security scanner, a content moderation system, a prompt injection firewall, and a security assessment toolkit. The project enables the identification and blocking of jailbreaking attempts and malicious prompts during model inference. It includes capabilities for detecting violating content across multiple languages and modalities and scanning generated code for vulnerabilities to prevent the execution of insecure commands. The framework further

NeMo-Guardrails is a toolkit for adding programmable safety constraints and dialogue boundaries to large language model conversational systems. It functions as security middleware that intercepts inputs and outputs to block prompt injections, jailbreaks, and sensitive data leaks, while providing a conversational dialogue manager to define structured interaction flows through configuration files. The framework includes a hallucination filter to screen model outputs for factual accuracy and a specialized modeling language for defining conversational flows and constraints. It provides capabiliti

This repository provides tools and methodologies for studying adversarial attacks on large language models. It focuses on understanding how carefully crafted inputs can manipulate or bypass the safety mechanisms of LLMs, enabling researchers to probe model vulnerabilities and improve their robustness. The project covers techniques for generating adversarial prompts, evaluating model responses under attack conditions, and analyzing the effectiveness of different attack strategies.

L1B3RT4S is an adversarial machine learning toolkit designed for red teaming and evaluating the robustness of large language models. It provides a research framework for investigating how safety alignment mechanisms and content moderation systems respond to sophisticated input strategies. The project focuses on identifying vulnerabilities in model guardrails by employing techniques such as adversarial narrative framing, dynamic context injection, and latent space steering. It utilizes multi-agent prompt decomposition and recursive text transformation to analyze how structural changes to input

RagaAI-Catalyst is a suite of software implementation tools providing an SDK, dashboard, and platform for monitoring, debugging, red-teaming, and evaluating agentic AI workflows. It serves as an observability framework for tracing the execution paths of large language models and multi-agent systems. The project distinguishes itself through a security suite for automated red-teaming and vulnerability scanning to detect biases, alongside a centralized prompt registry that decouples templates from application code. It further provides an evaluation platform that combines synthetic data generatio

Superagent is a framework for AI assistant orchestration and agent security. It provides the tools to build intelligent assistants that integrate external APIs and maintain conversation memory to automate complex tasks. The project focuses on AI agent security through adversarial testing, red teaming, and the detection of prompt injections and malicious tool calls. It includes automated vulnerability patching, which scans codebases and configurations for security flaws and generates pull requests with fixes. The platform supports retrieval augmented generation by connecting language models t

LLM Prompt Injection Detector

Cleverhans is an adversarial machine learning library and toolkit designed to generate adversarial examples, incorporate them into training loops, and benchmark the resilience of machine learning models. It provides a gradient-based attack framework for constructing both white-box and black-box attacks to identify model misclassifications. The project includes capabilities for model robustness benchmarking, allowing users to evaluate and verify how models resist evasion attacks and malicious input perturbations. It also facilitates adversarial training to increase a model's resistance to pert

A collection of real world AI/ML exploits for responsibly disclosed vulnerabilities

This project is no longer actively maintained. You are welcome to fork and continue its development on your own. Thank you for your interest and support.

The Adversarial Robustness Toolbox (ART) is an open-source library that provides a unified framework for evaluating, defending, and certifying machine learning models against adversarial threats. It wraps models from any framework behind a common estimator interface, enabling composable pipelines for attack generation, defense application, robustness certification, and privacy auditing across evasion, poisoning, and extraction threats. The library distinguishes itself by covering the full adversarial ML security lifecycle within a single toolkit. It supports gradient-based adversarial example

LLM Guard is a security firewall and guardrail framework designed to scan and sanitize inputs and outputs for large language models. It functions as a proxy gateway and security layer to block prompt injections, toxicity, and sensitive data leakage while ensuring that model interactions remain compliant with organizational policies. The system distinguishes itself through a modular scanner pipeline that utilizes local model orchestration to eliminate external network dependencies. It supports real-time security filtering via streaming chunk analysis and implements a fail-fast execution model

Snyk is an application security testing platform designed to identify and remediate vulnerabilities across source code, open-source dependencies, container images, and infrastructure-as-code configurations. It functions as a comprehensive security workflow automation tool, utilizing a static analysis engine and dependency graph mapping to detect security flaws and license compliance issues throughout the software development lifecycle. The platform distinguishes itself through agentic workflow orchestration and an automated remediation pipeline that generates and submits pull requests to patc

The automated security helper is a command-line utility designed to orchestrate multiple security analysis tools into a unified, configuration-driven workflow. It functions as a central engine that executes static application security testing and infrastructure scans, aggregating diverse tool outputs into a standardized, machine-readable format to ensure consistent vulnerability detection across development lifecycles. The tool distinguishes itself through a modular plugin architecture that allows for the integration of custom or proprietary scanners, alongside an external intelligence layer

RED_HAWK is a penetration testing framework and reconnaissance suite designed for information gathering and vulnerability assessment. It provides a toolkit for infrastructure reconnaissance, technology stack detection, automated web spidering, and security scanning. The project distinguishes itself through a multi-stage reconnaissance pipeline that maps attack surfaces. This includes DNS-based infrastructure mapping to resolve network layouts and pattern-based detection to identify specific content management systems and server stacks. The system covers a broad range of capabilities includin

This project is an AI-powered static analysis tool and automated vulnerability scanner designed to detect security flaws such as injection and authentication bypasses. It uses large language models to perform semantic reasoning across multiple programming languages, identifying vulnerabilities within code changes. The tool operates as a GitHub Action that integrates into continuous integration pipelines to analyze pull request diffs. It focuses on modified lines of code to target new risks and reports findings by posting automated comments directly to the pull request. Analysis is directed b

fuzzDicts is a repository of curated wordlists and dictionaries designed for web application fuzzing. It provides collections of strings and payloads used to discover hidden files, subdomains, and security vulnerabilities. The project includes specialized libraries for different security testing vectors, such as dictionaries for common request and cookie parameters, lists of common subdomain prefixes, and collections of passwords and default vendor credentials for brute-force testing. It also maintains a security payload library containing character sequences used to identify flaws like SQL i

Ghauri is an automated SQL injection scanner and exploitation tool designed to detect and extract data from vulnerable databases. It functions as a database exfiltration framework that identifies security flaws and retrieves system banners, hostnames, and database schemas. The tool identifies boolean, error, time-based, and stacked query vulnerabilities across multiple input vectors, including HTTP headers, cookies, JSON, SOAP, and XML. It provides capabilities for automated database exfiltration and the processing of bulk target lists to identify flaws across multiple environments. The syst

Maskphish is a comprehensive security toolkit that integrates capabilities for digital forensics, network vulnerability scanning, open-source intelligence, penetration testing, and social engineering. It functions as a multi-purpose framework for automating reconnaissance and executing security audits across diverse network environments. The project features a specialized phishing and social engineering toolkit used for cloning websites, masking URLs, and deploying deceptive pages to capture user credentials. It also includes a remote access Trojan builder for generating platform-specific exe

TscanPlus is an external attack surface management tool and security reconnaissance framework designed for discovering network assets, enumerating subdomains, and mapping internet-facing services. It functions as a vulnerability scanning framework and network asset discovery suite to identify security exposure and map active hosts. The platform distinguishes itself by integrating an intelligence layer that uses large language models to analyze raw scan results and identify security weaknesses within JavaScript code. It also includes a dedicated proxy management system that validates and rotat

Astra is a security analysis system and scanner designed to identify vulnerabilities and security flaws in REST API endpoints. It functions as a security testing tool that automatically detects common API weaknesses during development and deployment cycles. The project provides a graphical interface for triggering and monitoring security scanning processes, removing the requirement for manual command line execution. This management UI allows for the oversight of scanning workflows and the retrieval of vulnerability reports. The system supports the import of collection files to map endpoints

This project is a set of git pre-commit hooks designed to automate the formatting, linting, and validation of Terraform configurations. It functions as an infrastructure as code linter, security scanner, cost estimator, and documentation generator to ensure code quality before commits are finalized. The tool distinguishes itself by providing specialized capabilities for infrastructure workflows, such as scanning templates for security vulnerabilities and hardcoded secrets, calculating projected cloud spending against budgets, and automatically extracting module definitions to populate readme

This project is a comprehensive Android reverse engineering suite that functions as a decompiler, bytecode deobfuscator, and malware analysis tool. It is designed to convert APK, DEX, and OAT binaries into human-readable source code using a native implementation that does not require a Java Virtual Machine. The platform is distinguished by its integration with Frida for dynamic analysis, allowing users to hook methods, inject custom JavaScript, and dump device memory in real time. It also features specialized security engines, including a taint propagation engine and a stack-state machine, to

ShuiZe_0x727 is an open-source intelligence gathering framework and attack surface management tool. It functions as an asset discovery engine and cyber intelligence aggregator designed to identify internet-facing assets, map network infrastructure, and visualize total network exposure. The project integrates vulnerability scanning and sensitive data leak detection to identify security weaknesses and unauthorized access points. It employs a combination of network space API queries, certificate log analysis, and public repository scanning to extract leaked credentials, API keys, and internal ad