Search Results for "guard"

LLM Guard is an open-source security toolkit designed to protect large language model applications from various security risks and adversarial attacks. The library acts as a protective layer between users and language models by analyzing inputs and outputs before they reach or leave the model. It includes scanning mechanisms that detect malicious prompts, prompt injection attempts, toxic content, and other harmful inputs that could compromise AI systems.

AWorld (Agent World) is an agent runtime/framework. It supports building, evaluating, and training self-improving intelligent agents and multi-agent systems (MAS). It is designed to provide infrastructure for agent orchestration, iterative learning, and environment interaction at scale. Scalable training across environments and distributed setups. Support for multi-agent collaboration/orchestration (MAS). The system is intended to help agents evolve via experience. It provides features to...

AI-Infra-Guard is a powerful open-source security platform from Tencent’s Zhuque Lab designed to assess the safety and resilience of AI infrastructures, codebases, and components through automated scanning and evaluation tools. It brings together AI infrastructure vulnerability scanning, MCP server risk analysis, and jailbreak evaluation into a unified workflow so that enterprises and individuals can identify critical security issues without relying on external services.

...Its scope spans input and output safeguards, cybersecurity-focused evaluations, and reference shields that can be inserted at inference time. The project evolves as a hub for safety research artifacts like Llama Guard and Code Shield, along with dataset specs and how-to guides for integrating checks into applications. CyberSecEval, one of its flagship components, provides repeatable evaluations for security risk, including agent-oriented tasks such as automated patching benchmarks. The aim is to make safety practical: ship testable baselines, publish metrics, and provide drop-in implementations that reduce friction for teams adopting Llama. ...