vellaveto

What it does

Vellaveto is an agentic security control plane designed to protect AI agent environments by managing and monitoring tool/function calls through the Model Context Protocol (MCP). It acts as a policy gateway that intercepts and validates all agent tool invocations before execution, implementing security controls like data loss prevention (DLP), compliance enforcement (EU AI Act), and OAuth integration.

Key features

• MCP-native policy gateway for agent tool calls
• Automatic topology discovery for agent environments
• Comprehensive audit logging and SIEM integration
• Data loss prevention (DLP) and compliance controls
• Policy engine with customizable security rules
• OAuth support for authentication and authorization
• OWASP-aligned security standards

How to set up

Vellaveto is deployed as a gateway that sits between your AI agents and MCP servers. Installation involves configuring it as an MCP proxy, defining your security policies, and connecting it to your audit/SIEM systems. Detailed setup documentation is available on the GitHub repository.