ModelRed

ModelRed is an AI security & red-teaming platform that continuously tests and stresses LLMs and agents (via probes, SDK, CI/CD integration) for vulnerabilities such as prompt injection, data leaks and risky tool calls.