Aegis — AI Agent Governance

Aegis — AI Agent Governance is an MCP (Model Context Protocol) server that provides a universal governance and security layer for AI agent tool calls across 12+ frameworks, including LangChain, CrewAI, OpenAI, Anthropic, LiteLLM, and more. It delivers YAML-based policy enforcement, approval gates, PII masking, toxicity detection, audit logging, prompt injection guardrails, and secure, tamper-evident audits. Aegis is aimed at developers, security engineers, and organizations seeking to ensure compliance, security, and trust for autonomous AI agents in production by standardizing security primitives across frameworks with minimal integration effort.