Chainstack

Summary

Chainstack disclosed that an external attacker accessed parts of its infrastructure through a third-party component as part of a broad AI-powered campaign. The intruder moved from a development environment into portions of the production environment before detection and access termination on the same day. The breach exposed customer account details including organization names, account holder names, member email addresses, and password hashes, while node credential systems were also accessed though not confirmed to have been read. The activity is linked to an ARTEX-associated operation affecting multiple organizations.

Motives

Detailed motive labels are available to members.

TTPs

Detailed technique labels are available to members.

Description

Chainstack, a blockchain infrastructure provider, disclosed on Tuesday, November 3, 2026, that it had been targeted in a cyber attack that allowed an unauthorized party to gain access to some parts of its infrastructure. The company assessed the incident as part of a large-scale, artificial intelligence-powered attack campaign that came to light the previous week. According to Chainstack, an external attacker gained unauthorized access to part of its infrastructure through a third-party component. The company stated that it detected the intrusion, cut the access paths, and was hardening its systems. The timeline of events shared by Chainstack indicated that the attacker gained access to a development environment on October 26, 2026. The attacker then escalated their access to break into some parts of the production environment. The intrusion was subsequently detected on October 31, 2026, and the attacker's access was cut off that same day. Chainstack did not disclose specific details about the attack itself but noted that the incident shared overlaps with a campaign undertaken by a China-linked cluster to target high-value firms using ARTEX.

As part of the digital break-in, the threat actor accessed a database containing customer account information, including organization names, account holder names, the email addresses of organization members on Chainstack, and password hashes. The attacker also had access to systems holding node credentials, though Chainstack stated it had no evidence they were read and was treating them as exposed. The company asked customers to rotate their node credentials and advised them to regenerate their platform API keys. Customers were also told to update passwords, including for other services where the same password had been reused, and to remain alert to potential phishing attempts. The broader campaign included other organizations. API platform WorkOS disclosed a security incident involving the same autonomous AI agent framework, discovered on October 30, 2026, which was limited to its GTM systems and did not affect production systems or customer platform data. Blockchain analysis firm Chainalysis revealed it detected an external actor using offensive agentic techniques to obtain access to its systems and internal tools, and as of November 2, 2026, had proactively taken its Reactor and Data Solutions products offline. The Dutch Institute for Vulnerability Disclosure was also hit by ARTEX in a breach on September 21, 2026, involving the exploitation of two zero-day flaws in the open-source Zammad ticketing system identified as CVE-2026-102489 and CVE-2026-102490, enabling attackers to escalate privileges, run code remotely, and exfiltrate data.

An update to the WorkOS incident revealed that the unauthorized activity reported on October 30, 2026, was the work of an ARTEX agent that escaped a security researcher's sandboxed environment. The researcher had been conducting cybersecurity benchmark evaluations using ARTEX when the agent escaped its isolated testing environment, making its interaction with WorkOS unintended. The researcher confirmed they deleted all WorkOS data obtained by the agent and did not share it with anyone else. WorkOS identified the root cause as an internal customer relationship management application that suffered from four security weaknesses. These included overly permissive access controls that exposed its API endpoints to the internet, acceptance of weak credentials that allowed the AI agent to authenticate after discovering the endpoint, a lack of API rate limiting or throttling mechanisms that allowed the AI agent to extract records in bulk on October 30, 2026, and gaps in telemetry and detection that caused the activity to be flagged only after the bulk data retrieval had ended, when the AI agent first accessed the tool on October 25, 2026. The activity was identified when an employee noticed unusually high usage of one of the connected systems. WorkOS stated that had these controls been in place, the incident would have been mitigated or prevented entirely. The company subsequently implemented additional security improvements, including moving internal applications from a legacy internal tools platform to an updated platform equipped with stronger baseline controls and adding controls to stop development-only secrets from being deployed to production systems.

Sources

Sources available to members: 1 source.