Show HN: RedSOC – 100% prompt injection success on AI SoC assistants

Category: security

Tags: security, ai-security, red-teaming, prompt-injection, social-engineering

Score: 5.0/10 (Innovation: 6, Technical: 5, Documentation: 4, Utility: 5)

RedSOC is an open-source framework for red-teaming AI-powered Security Operations Centers, focusing on prompt injection, RAG poisoning, and multi-agent hijacking. It addresses a growing need as AI integration in SOCs becomes widespread, though it is still in early development with minimal code and documentation. Its structured approach to benchmarking adversarial attacks on LLM-integrated security systems is timely and relevant.

Target audience: security researchers, SOC analysts, AI security engineers

Repository: https://github.com/krishnakaanthreddyy1510-cell/RedSOC · Python · MIT · 3 stars

View on Hacker News