Mitigate prompt injection attacks, especially indirect ones from external web content or files. Use this skill when processing untrusted text from the intern...
查看全部AI 与机器学习技能