# Red Teaming

UK AI Security Institute framework for LLM safety evaluations

Inspect AI is an MIT-licensed framework from the UK AI Security Institute for running large language model evaluations, including tool use, multi-turn dialogue, model-graded scoring, and reusable evaluation tasks.

Microsoft’s pytest-native red teaming framework for turning AI agent safety findings into CI tests.

RAMPART is an open-source Microsoft framework for safety and security testing of agentic AI applications. It brings red-team findings into a pytest-native workflow so teams can turn prompt injection, unsafe tool use, and behavioral boundary failures into repeatable regression tests. The strongest aicoolies angle is developer workflow: RAMPART makes agent safety part of CI/CD instead of a one-off security review.

LLM vulnerability scanner and red teaming kit

Agentic Security is an open-source vulnerability scanner for LLM agent workflows that tests AI systems against jailbreaks, fuzzing, and multimodal attacks. It probes weaknesses across text, image, and audio inputs through multi-step jailbreak simulations, randomized stress testing, and reinforcement learning-powered adaptive attacks. The toolkit connects directly to LLM APIs for high-volume real-world attack scenarios, helping developers identify and patch safety gaps before deployment.

Microsoft's automated red teaming framework for AI systems

PyRIT (Python Risk Identification Toolkit) is Microsoft's open-source framework for automated red teaming of generative AI systems. It enables security researchers to probe LLMs for jailbreaks, prompt injection, content safety bypasses, and harmful output generation using multi-turn attack strategies, scoring engines, and orchestrated adversarial workflows. Supports multiple target models and integrates with Azure AI services.

AI agent safety SDK with guard, redact, and scan modules

Superagent is an open-source AI agent safety SDK that provides runtime protection through four modules: Guard for detecting prompt injections and unsafe tool calls, Redact for removing PII and secrets, Scan for analyzing repos against AI-targeted attacks, and Test for red-team evaluations. It works with any LLM provider and includes open-weight guard models from 0.6B to 4B parameters with 50-100ms latency for real-time protection.

Open-source diagnostic for AI operational misalignment

iFixAi is an Apache-2.0 diagnostic tool for scoring AI agents and models against operational-misalignment risks such as hallucination, manipulation, sabotage, sandbagging, and oversight evasion.