Self-contained MCP security server that scores AI agent tool-calls with a rule-based library + behavioral detector, giving explainable, MITRE-mapped verdicts for known attacks
python agent mcp cybersecurity security-tools detection-model ai-security ai-tools cybersecurity-tools mitre-atlas cybersecurity-projects prompt-injection llm-security owasp-llm-top-10 agent-security mcpgateway mcp-security mcp-vulnerability-scanner tool-call-validation
-
Updated
Jul 16, 2026 - Python