I think this service is already pretty good, but when defending against LLM based attacks, they're a godsend. Current tools don't appear to be too discerning.

Thanks! We also think they’re one of the few things available that work well for agentic threats.

You might find some recent work we did on agent deception interesting as well: https://blog.thinkst.com/2026/09/getting-agents-to-tell-on-t...