Shypd.aiThe largest open index of AI toolsagentic_security is an open-source AI red teaming kit and LLM vulnerability scanner that helps protect AI systems from jailbreaks, fuzzing, and multimodal attacks. It allows security engineers to perform automated security testing.
Trending
About
agentic_security is an open-source vulnerability scanner and AI red teaming kit designed to safeguard Large Language Models (LLMs) and agent workflows against emerging threats. It provides powerful tools for security teams, developers, and researchers to proactively identify and mitigate risks in AI systems, ensuring more reliable and secure deployments. Key features include the ability to probe vulnerabilities across text, images, and audio inputs for multimodal attacks, simulate sophisticated multi-step jailbreaks, and stress-test LLMs with comprehensive fuzzing using randomized inputs. The tool also offers seamless API integration for stress testing with high-volume, real-world attack scenarios and leverages reinforcement learning to craft adaptive, intelligent probes that evolve with model defenses. Installation is straightforward via pip, and it supports custom datasets and CI/CD integration.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
