Comment

Comment by ipince

This seems to prevent your keys from being exfiltrated through prompt injection. But if your agent could've been prompt injected into giving out keys, then it can also be prompt injected into using the services it has (fake) keys for to the attacker's benefit.
parent ipince 7 months ago

No comments yet

Be the first to share your thoughts.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton