Show HN: AegisKernel – Test prompt injection attacks on AI agents in the browser

Category: security

Tags: prompt-injection, ai-security, browser-based, zero-trust, testing

Score: 4.7/10 (Innovation: 5, Technical: 4, Documentation: 2, Utility: 5)

AegisKernel is a browser-based environment for testing prompt injection attacks against AI agents, featuring a zero-trust interceptor runtime. It allows users to simulate attack vectors, but the project is in an early prototype stage with minimal documentation.

Target audience: AI developers, security researchers

Repository: https://adikshatriyag-crypto.github.io/aegis-kernel/

View on Hacker News