AIProxyGuard
Block Prompt Injection Before It Matters
A security layer that scans every LLM request and response for malicious patterns. Route across OpenAI, Anthropic, and Ollama with per-category policies and Prometheus metrics.
Get Started (opens in new tab)