SPILLWAY: deterministic defense stops all templated prompt injections on tool-using agents

saintmachines · hf · 2026-11-17

A paper from SAINT Machines, KAIST, and Chung-Ang University introduces SPILLWAY, a defense for tool-using agents against prompt injection.

An auditable refusal floor for personal agents handling email and payments.

Original post →

More from Safety

Safety channel →