Cyber defence is being automated: Positive Technologies opens its products to AI agents

AI agents will be able to collect data from various systems, analyse threats, investigate incidents, and participate in attack response

Positive Technologies will add application programming interfaces for interacting with AI agents across all its products. They will make it possible to automate work with security tools: collecting and analysing data, investigating incidents, and launching response scenarios.

Image source: Grok Imagine AI

For integration, the company released MCP servers based on the open Model Context Protocol. With their help, AI agents will be able to access the tools of individual products and use data from several systems simultaneously. By the end of 2026, technology support is planned to be implemented across the entire product lineup, including MaxPatrol SIEM, PT NAD, MaxPatrol Endpoint Security, MaxPatrol VM, and PT Sandbox.

A specialist will be able to describe a task in natural language, after which the agent will independently select the required tools. For example, it will be able to combine event data from SIEM with network telemetry, reconstruct the incident picture, and prepare response measures, including isolating an infected device.

The company attributes the launch to the growing number of AI-driven attacks. According to PT AF Cloud, in Q3 2026, the number of security events showing signs of using such technologies increased by 65% compared to the previous quarter.

MCP servers will be deployed inside the customer’s infrastructure and access products via the internal network. Agent access will be restricted by the permissions of the issued tokens. Operation in isolated environments with local language models is also supported.