Search Results: context-manipulation

prompt-guard

Detect and neutralize prompt injection attacks in OpenClaw skill content, user inputs, and external data sources. Prevents instruction hijacking and context manipulation.