Onyx Security

Company Overview

Snapshot

Founded in June 2024 by Maxim Kogan, Gil Elbaz, and Shachak Zicher, Onyx Security operates with 11–50 employees. The company has raised $40M across two funding rounds from three investors.

Business overview

Onyx Security provides an AI agent security platform designed for enterprise security teams. The platform's core technology enables the discovery and inventorying of AI agents across an organization, continuously assessing their access to systems and data. It allows security teams to define and enforce policies for AI agent behavior with real-time guardrails. The Onyx Guardian Agent monitors agent runtime behavior, identifies excessive permissions, blocks unsafe actions, and assists in incident response, aiming to facilitate the secure adoption of agentic AI at scale within enterprises without compromising visibility, control, or compliance. The company operates within the Cyber Security sector, serving enterprise and professional services markets.

Strategic signal

This investment signals a growing market recognition of the critical need for robust security solutions as enterprises increasingly adopt agentic AI. For investors, this indicates strong validation of Onyx Security's approach to providing control and oversight over AI agents, positioning the company for accelerated development and market penetration in the evolving cybersecurity landscape.

Company Intelligence Q&A

What is Onyx Security's primary offering?

Onyx Security provides an AI agent security platform that helps enterprise security teams discover, govern, and protect all AI agents within their organization. The platform builds an inventory of AI agents and continuously assesses their access to systems and data.

When was Onyx Security founded and by whom?

Onyx Security was founded in June 2024 by co-founders Maxim Kogan, Gil Elbaz, and Shachak Zicher.

What is the function of the Onyx Guardian Agent?

The Onyx Guardian Agent monitors AI agent behavior at runtime, flags excessive permissions, blocks unsafe actions, and supports incident response, ensuring enterprises can adopt agentic AI at scale while maintaining control and compliance.