· via dev.to (home feed)
FBI FLASH advisory: AI-driven ARTEX framework takes intrusions to privileged access in under three hours
The FBI says threat actors are using the open-source ARTEX framework to run autonomous, multi-target intrusions that reach privileged access in under three hours, including in the ConnectWise Cloud Backup supply-chain attack.

What the FBI says
The FBI has issued a FLASH advisory warning that threat actors are weaponizing ARTEX, an open-source, AI-driven penetration testing framework, to run automated intrusions that move from initial targeting to privileged access in under three hours. According to a RedEye Threat Intelligence write-up republished on dev.to, the advisory was released on Friday, November 14, and reports that multiple groups have used the tool since at least August 2026 against organizations in the financial, government, technology, defense and healthcare sectors.
The bureau's central concern is scale rather than novelty: ARTEX reportedly allows operators with little technical skill to run complex operations against several organizations at once, cutting the time each intrusion takes and widening how many targets a single person can hit.
How ARTEX works
ARTEX is written in Go and uses AI agents to handle reconnaissance, vulnerability discovery, exploitation, privilege escalation and data theft with minimal human input. It does not ship its own model; it connects to external large language models, including Anthropic Claude, OpenAI GPT, DeepSeek, GLM, Grok and Kimi. The operator states goals and tasks, and the framework breaks them into commands executed by specialized agents.
The FBI notes ARTEX is particularly effective against systems defenders often overlook: partner portals, employee-facing applications, legacy services and third-party integrations.
The developer, who uses the name Autumn-27, published ARTEX on GitHub in July 2026 and deleted the repository on October 8 after abuse reports. The code still circulates through community forks, archived copies, localized versions and a paid edition marketed as ARTEX PRO. RedEye reports that researchers counted more than 1,600 publicly exposed ARTEX consoles in early November, with true deployment numbers believed to be higher.
The ConnectWise supply-chain attack
The most significant campaign tied to ARTEX in the advisory is October's compromise of ConnectWise Cloud Backup, formerly SkyKick Cloud Backup. The attackers compromised the application identities the backup service used to reach customer Microsoft 365 environments. Because customers had already granted those identities permissions, the attackers could authenticate to tenants as a trusted application without phishing a single user.
Microsoft attributed the campaign to Storm-0940, a China-state-aligned actor, and its investigation found the group used ARTEX to speed up reconnaissance, password spraying and abuse of cloud application permissions. After hijacking the ConnectWise identities, the group attempted authentication against more than 2,250 downstream tenants, with some intrusions reaching credential modifications and persistence attempts. Huntress reported at least 162 compromised tenants at the time.
South Korean financial sector campaign
The FBI also cites attacks on South Korean financial institutions that exposed personal information on roughly 67,000 people. In that campaign, a Chinese-speaking threat actor paired ARTEX with Claude Code agents. Together, the two campaigns show the framework being adopted both by state-aligned groups running supply-chain operations and by financially motivated actors pursuing regulated data.
Guidance and detection
There is no patch, since ARTEX is an attack tool rather than a vulnerable product. The FBI's 13-page advisory includes indicators of compromise, such as IP addresses, command-and-control infrastructure and suspicious network activity, along with hardening guidance. The recommendations focus on securing internet-facing applications including partner and employee portals, restricting access to cloud administration interfaces, reviewing third-party application permissions and authentication logs, rate limiting sensitive APIs, and monitoring for bursts of automated requests, unusual cloud activity, unexpected credential or privilege changes, and large outbound data transfers.
On the same day, researcher MalwareMustDie released an open-source ARTEX detection signature pack covering suspicious network requests, identity abuse, cloud activity, malicious files and automated attack patterns, plus a playbook that grades indicators by confidence. Both the FBI and the playbook caution that infrastructure indicators such as IP addresses go stale quickly and can generate false positives, while behavioral signals — request bursts, identity abuse by application principals, rapid privilege changes and bulk data movement — remain useful longer. Organizations that find ARTEX-related activity are encouraged to report it to the FBI or IC3.
Why it matters
ARTEX is not a new exploit; it changes the economics of intrusion. Reconnaissance and attack chaining that once required a skilled team over days can now run unattended across many targets within hours, which compresses the defender's response window below what many organizations can sustain — an alert queued overnight may arrive after the breach. The ConnectWise case shows where this hurts most: third-party application identities holding broad, rarely reviewed permissions into Microsoft 365 tenants. With the code now impossible to withdraw and commercial variants circulating, organizations should inventory which integrations hold standing access to their environments and test whether their detection-to-response time fits inside a three-hour window.
- #ai
- #cybersecurity
- #fbi-advisory
- #penetration-testing
- #supply-chain-attack
- #microsoft-365