AgentShield Guard vs prompt-injection-detector

Runtime gateway for detecting prompt injection and jailbreak attempts in LLM agents.

Prompt Injection Detector is a stateless security microservice built for the agent economy. Before your agent acts on any input — whether from a user, an email, a document, or another agent — run it through this scanner first. Detects: jailbreak attempts, instruction overrides, role-play manipulatio

Models

MCP Servers

—

—

—

—

—

—

—

—

—

—

37

0