The use of a new AI-powered offensive security tool by cyber threat actors appears to be spreading well beyond the campaigns first identified.

This tool, called ARTEX, is an open-source, AI model-driven, autonomous penetration testing framework released on GitHub by a user known as ‘Autumn-27’ in July 2026.

While ARTEX has already been integrated in several known malicious campaigns since October, a new report by internet traffic scan provider Censys estimated its use could be much more widespread than previously thought.

ARTEX's Growing Use by Cyber Threat Actors

Written in Go with an embedded Next.js web console, ARTEX offers an offensive command-and-control (C2) platform that communicates with several AI agents and can be plugged in to large language models (LLMs).

In October, two threat intelligence companies attributed its use as part of threat actor operations.

First, in early October, CrowdStrike linked ARTEX to a campaign against South Korean financial institutions. The analysts found ARTEX was used alongside Anthropic’s Claude AI model during the campaign, which ran from late September to early October 2026. CrowdStrike assessed with moderate confidence that the attacker is a Chinese speaker and financially motivated.

Later in October, Microsoft linked ARTEX to a large-scale compromise of ConnectWise customer tenants which led to IT management software provider taking a number of its cloud services offline. Microsoft attributed the campaign to Storm-0940, a China-state-aligned cyber threat actor also known as CovertNetwork-1658 or ORB07.

More recently, on November 4, the Indian Computer Emergency Response Team (CERT-In) warned financial and cryptocurrency organizations that a threat group tracked as ‘Crafty Falcon’ had been targeting them in successive attacks using several AI-powered tools, including ARTEX.

The original GitHub repository for ARTEX was taken down by its creator in October 2026 due to widespread threat actor abuse. The author claimed they were not involved in the ARTEX-linked attacks and that the tool was only meant for authorized penetration testing.

Nevertheless, the framework continues to proliferate globally via community forks, including localized variations such as ScopeWeaver and a commercial version dubbed ARTEX PRO.

In a separate report, Check Point researchers showed AI agents in ARTEX control panels are running large-scale attacks with little human input for just a few hundred dollars, though they still fail and hallucinate successes.

Censys Uncovers ARTEX's Global Footprint

A report published on November 5 by Silas Cutler, principal security researcher at Censys Advanced Research Collective (ARC), has provided further insights into ARTEX's use and the infrastructure surrounding it.

The researcher found that 1465 hosts worldwide are exposing an ARTEX console to the public internet. About 650 of these hosts sit within two blocks of identically configured hosting, leaving roughly 815 distinct-looking deployments.

This suggests that ARTEX is not simply appearing on a handful of isolated systems, but across hundreds of seemingly distinct deployments worldwide, indicating a much broader footprint than initially reported.

Additionally, Censys telemetry showed that 155 internet-exposed ARTEX hosts (10.6%) also run at least one other artefact categorized by Censys as a threat.

These include other AI attack frameworks, reconnaissance platforms, tunneling tools and traditional malicious command-and-control (C2) servers such as Vshell, Cobalt Strike and Adaptix C2.

An asset-discovery and vulnerability-scanning platform dubbed ScopeSentry, which is maintained by the creator of ARTEX, is also listed by Censys as a threat artefact.

The researchers also found ARTEX sharing infrastructure with other popular AI-powered penetration-testing tools, including CyberStrikeAI, PentAGI, HexStrike AI and RedAmon. CyberStrikeAI, in particular, appeared alongside ARTEX on 70 hosts.

"Operators appear to be using multiple agent frameworks side by side," Cutler suggested.

The ‘Crafty Falcon’ campaign reported by CERT-In in November showed usage of ARTEX alongside PentAGI and CyberStrikeAI.

Censys also identified 32 ARTEX hosts with publicly accessible directories containing tools used in different stages of cyber-attacks, including post-exploitation kits, stagers and implants.

The findings also point to the growing role of AI infrastructure in these operations and 43 ARTEX hosts were found to be running self-hosted LLM gateway or reseller software. This suggests that access to AI models is becoming another component of the offensive infrastructure stack.