This is a security auditing tool for your MCP infrastructure. It scans other MCP servers to identify vulnerabilities like prompt injection attacks and tool poisoning, where malicious actors could manipulate prompts or compromise tool definitions. Reach for this when you're running third-party MCP servers or building security-critical applications and need to verify that your MCP ecosystem isn't exposing attack vectors. Given the novelty of MCP as a protocol, having a dedicated scanner that understands MCP-specific threat models makes sense. The streamable HTTP transport means you can run security checks remotely without installing anything locally.
claude mcp add --transport http io.github.mcp-fortress-mcp-fortress https://server.smithery.ai/@mcp-fortress/mcp-fortress-server/mcp