Open-source agent security framework. Detects and defends against AI Agent Traps — content injection, embedded jailbreaks, RAG poisoning, data exfiltration, and more. Based on the DeepMind Agent Traps taxonomy.
AI model health monitor for LLM apps – runtime checks for drift, hallucination risk, latency, and JSON/format quality on any OpenAI, Anthropic, or local client.
A TypeScript library for validating and securing LLM prompts — PII, jailbreak, prompt injection, profanity, toxicity and relevance guards. Zero-config defaults, extensible with custom patterns.