Owasp Agentic Mcp

This server brings OWASP's Top 10 for Agentic AI security checks into your MCP workflow. It scans for prompt injection attempts, tool poisoning vulnerabilities, excessive agency issues, and data leakage risks in your AI agent interactions. Built by MEOK AI Labs with EU AI Act compliance hooks baked in, including risk management tracking and bias detection tooling. The core offering is free and MIT licensed, though they pitch enterprise tiers for SLA guarantees and custom development. You'd reach for this when you need to audit agent behavior against OWASP security standards or when regulatory compliance documentation matters for your deployment.