Is this text a prompt-injection trap for your AI agent? Checks a web page, email, tool result, API answer or another agent's message before your agent acts on it: does it try to make the agent ignore its instructions, pay or send money, hand over keys or private data, or obey a fake user, admin or support message? Green/orange/red with reason codes, judged by TypeSafe's Jev, plus hidden-character tricks and OFAC-sanctioned addresses in the text

Payment Options