Archestra's Dual LLM Pattern: Using "Guess Who?" Logic to Stop Prompt Injections

Re:Archestra's Dual LLM Pattern: Using "Guess Who?" Logic to... by joeyorlando on Monday October 13, 2025 @10:18AM (#45568598)

hey there

Joey here from Archestra. Good question. I recently was evaluating what you mention, against the latest/"smartest" models from the big LLM providers, and I was able to trick all of them.

Take a look at https://www.archestra.ai/blog/what-is-a-prompt-injection which has all the details on how I did this.

[ Parent | on HN ]

Re:Archestra's Dual LLM Pattern: Using "Guess Who?" Logic to... by magicalhippo on Monday October 13, 2025 @11:18AM (#45569287)

Thanks. Interesting and scary such blatant attempts succeed. After all, all external data is evil, we all know that right?

[ Parent | on HN ]

Re:Archestra's Dual LLM Pattern: Using "Guess Who?" Logic to... by ildari on Monday October 13, 2025 @11:42AM (#45569614)

external data is unavoidable for the properly functioning agent, so we have to learn to cook it

[ Parent | on HN ]

Re:Archestra's Dual LLM Pattern: Using "Guess Who?" Logic to... by magicalhippo on Tuesday October 14, 2025 @01:42AM (#45576655)

True, however this seems like such basic stuff. Download arbitrary text and inject it into your prompt?

Why on earth would you not consider that as a very dangerous operation that needs to be carefully managed? It's like parking your bike downtown hoping it wont be stolen. Like, at least use a zip tie or something.

That said, I agree with your post that this won't catch everything. So something else, like a quarantined LLM like you suggest is likely needed.

However I just didn't expect such blatant attacks to pass.

[ Parent | on HN ]