agent-input-sanitizer
Defend an agent against hidden-content injection: strip payload-capable invisible Unicode and ANSI, splice out human-invisible HTML, and flag data-exfil URLs in untrusted text before any model sees it.
Version 1.47.14 License Apache-2.0
Keywords
llmprompt-injectionsanitizeunicodeinvisible-charactersansiexfiltrationragagentsecurityhomoglyphconfusablesredactiontool-outputedit-repair
INSTALL