AI Agent Security Scanner

Overview

AI Agent Security Scanner is a Model Context Protocol server that scans AI agents and MCP servers for common security vulnerabilities including injection attacks, jailbreak attempts, and tool abuse. Once integrated, it exposes security analysis capabilities as tools that your AI assistant can invoke directly.

This is a remote-hosted server requiring no local installation. Point your MCP client to https://llmrt-companion.manhliemcn4euwlu.workers.dev/mcp-agent-security to connect immediately. The current release is v1.1.0, and no authentication is required.

AI Agent Security Scanner is listed in the AI & Agents category and works with Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, and other MCP clients. If you maintain this server, claim your listing to verify ownership and keep information current.

Installation

Settings → Connectors → Add custom connector
Name: AI Agent Security Scanner
URL:  https://llmrt-companion.manhliemcn4euwlu.workers.dev/mcp-agent-security