RedLockX Update: New Website & Interactive Research Hub We've just launched a major update to RedLockX!
To make our project easier to explore for judges, researchers, and developers, we've built a dedicated project website that showcases the complete RedLockX ecosystem in one place.
What's New? • Interactive landing page with live cybersecurity-inspired visuals • Detailed breakdown of the Prompt Injection threat landscape • Real-world AI security incidents and CVEs • Visual architecture of our dual-model detection pipeline • Attack gallery covering jailbreaks, prompt leaks, obfuscation, and indirect injections • Live Hugging Face integration so visitors can test RedLockX directly • Performance metrics, research insights, and detection methodology
Our goal is not only to detect prompt injection attacks, but also to help raise awareness about one of the fastest-growing threats facing AI systems today.
We're continuing to refine the platform and expand RedLockX into a complete security layer for LLM-powered applications. Feedback is always welcome. Stay tuned for more updates!
Log in or sign up for Devpost to join the conversation.