EXPLORE

Article

Vulnerability Discovered in Anthropic’s Claude Code

Arete Analysis

Security researchers discovered two critical vulnerabilities in Anthropic's agentic AI coding tool, Claude Code. The vulnerabilities, tracked as CVE-2025-59536 and CVE-2026-21852, allowed attackers to achieve remote code execution and to compromise a victim's API credentials. The vulnerabilities exploit maliciously crafted repository configurations to circumvent control mechanisms. It should be noted that Anthropic worked closely with the security researchers throughout the process, and the bugs were patched before the research was published. 

What’s Notable and Unique 

  • The configuration files .claude/settings.json and .mcp.json were repurposed to execute malicious commands. Because the configurations could be applied immediately upon starting Claude Code, the commands ran before the user could deny permissions via a dialogue prompt, or they bypassed the authentication prompt altogether. 


  • .claude/settings.json also defines the endpoint for all Claude Code API communications. By replacing the default localhost URL with a URL they own, an attacker could redirect traffic to infrastructure they control. Critically, the authentication traffic generated upon starting Claude Code included the user's full Anthropic API key in plain text and was sent before the user could interact with the trust dialogue. 


  • Restrictive permissions on sensitive files could be bypassed by simply prompting Claude Code to create a copy of the file's contents, which did not inherit the original file's permissions. A threat actor using a stolen API key could gain complete read and write access to all files within a workspace. 

Analyst Comments 

The vulnerabilities and attack paths detailed in the research illustrate the double-edged nature of AI tools. The speed, scale, and convenience characteristics that make AI tools attractive to developer teams also benefit threat actors who use them for nefarious purposes. Defenders should expect adversaries to continue seeking ways to exploit configurations and orchestration logic to increase the impact of their attacks. Organizations planning to implement AI development tools should prioritize AI supply-chain hygiene and CI/CD hardening practices. 

Sources 

Back to Blog Posts

Article

Europol Disrupts AudiA6 Crypto Laundering Service

European authorities have dismantled AudiA6, a major cryptocurrency laundering service linked to ransomware groups and broader cybercriminal networks. Between 2022 and 2025, the platform is believed to have processed over €336 million in illicit funds, enabling threat actors to obscure financial trails and monetize cybercrime proceeds. Its operators are also suspected of running Dark2Web, a dark web forum that facilitated collaboration, services, and connections among cybercriminals globally. This development underscores the expanding role of sophisticated, large-scale cryptocurrency laundering services in sustaining the cybercrime economy, enabling threat actors to obscure illicit funds and evade regulatory controls.

What’s Notable and Unique 

  • Following law enforcement disruption of Cryptex and Garantex, AudiA6 emerged as another platform involved in financial activities linked to ransomware groups. Investigators believe that AudiA6 became a central hub for cybercriminals seeking to launder stolen digital assets while obscuring the transaction trail from authorities.

  • On June 10, 2026, a coordinated operation resulted in two arrests in Georgia, the dismantling of key infrastructure (30+ servers, 25 domains), the freezing or seizure of over €778,000 in crypto, and the takedown of the AudiA6 and Dark2Web platforms. 

Analyst Comments

Ransomware groups and cybercriminal networks are increasingly leveraging sophisticated techniques, including chain-hopping, decentralized exchanges, and mixer-as-a-service platforms, to rapidly move illicit cryptocurrency across multiple blockchains, effectively obscuring transaction trails. Concurrently, the widespread use of fraudulent exchange accounts, mule wallets, and privacy-enhancing tools has elevated cryptocurrency laundering to a core enabler of the cybercrime ecosystem, allowing actors to bypass anti-money-laundering controls at scale. This investigation identified over 6,000 KYC records linked to money-mule accounts, many of which were tied to Russian-speaking intermediaries specifically recruited to facilitate the movement of illicit proceeds. These threat actors systematically used both commercial and domain-controlled email services to establish mule accounts across multiple cryptocurrency platforms. Collectively, these findings underscore the growing scale, coordination, and professionalization of cryptocurrency-enabled crime, highlighting the critical need for sustained, intelligence-led, and internationally coordinated efforts to disrupt these evolving financial ecosystems.

Sources

  • Ransomware gangs cut off from EUR 336 million ‘AudiA6’ crypto laundering pipeline

Article

Threat Actors Leverage AI for EDR Evasion

A threat actor has developed and deployed a ransomware attack toolkit enhanced with AI-assisted development workflows, enabling automated Active Directory (AD) discovery and improved EDR evasion capabilities. The toolkit leverages agent-based AI systems, such as Claude’s Opus and Cursor agents, for iterative malware development, testing, and refinement. 

What’s Notable and Unique 

  • Researchers have highlighted that this toolkit can not only generate ransomware code but also bypass sophisticated security defenses and identify AD networks for malware distribution. 

  • The framework incorporates multiple capabilities, including automated AD discovery and reconnaissance mechanisms, iterative EDR testing environments to refine evasion techniques, and a command-and-control (C2) infrastructure that leverages Telegram APIs and Cloudflare redirectors for stealth. 

  • Additionally, some agents were tasked with checking security research and technical posts for various bypass techniques. The agents recognized what was required for reproduction, extracted the techniques, mapped them to the MITRE ATT&CK knowledge base of adversary behaviors, set up a test lab, carried out the methodology, and reported the results. 

  • After a few repetitions, the modules seemed to avoid nearly all EDR solutions, despite the agent’s initial suggestion of a high failure rate. Although researchers found no evidence that AI was embedded in deployed malware or was operating independently in victim environments, the technology was still used to accelerate the iterative process of developing, testing, and refining payloads against security products, shortening the period between the publication of offensive security research and its practical implementation by threat actors. 

Analyst Comments 

AI-driven tools like this could accelerate the pace and sophistication of ransomware attacks, enabling even relatively inexperienced actors to launch high-impact campaigns. This development underscores the urgent need for security solutions to adapt to AI-assisted threats. Organizations must respond by strengthening detection engineering, improving visibility across environments, and maintaining robust security fundamentals.  

Sources 

  • AI-built ransomware toolkit automates EDR evasion, AD discovery  

  • Pointing a Cursor at evading detection

Article

Arete's 2026 Q1 Crimeware Report

Harness Arete’s unique data and expertise on extortion and ransomware to inform your response to the evolving threat landscape.

Article

CMS Vulnerability Leads to ClickFix Campaign

Threat actors compromised at least 700 education and technology websites in a recent ClickFix campaign by exploiting a critical SQL injection flaw (CVE-2026-26980) in the Ghost content management system (CMS). Adversaries combined the vulnerability with the ClickFix social engineering tactic to steal admin keys and inject a malicious JavaScript that delivers a fake Cloudflare or CAPTCHA verification pop-up, tricking victims into copying and pasting a malicious command into their systems.

What’s Notable and Unique

  • Rather than targeting the end user first, this campaign is unique in its initial exploitation of the system, followed by social engineering attempts. This hybrid attack style is likely being leveraged to bypass traditional defenses.

  • This recent campaign also highlights how trusted web properties can be weaponized at scale and coupled with unpatched CMS vulnerabilities. Rather than using the CMS compromise to perpetrate a single attack, threat actors turned it into a supply-chain attack that ultimately affected over 700 trusted websites.

Analyst Comments

As network defenders and their tools enhance threat detection capabilities, adversaries increasingly seek methods to bypass these defenses. By combining vulnerability exploitation, social engineering techniques, and staging for ancillary attacks, this campaign successfully bypassed traditional defenses and inflicted significant impact. Defending against hybrid cyberattacks requires comprehensive security controls beyond simply patching vulnerabilities. Organizations should focus on limiting movement within the environment, detecting abuse of trusted applications, and preventing end-user manipulation.

Sources

  • 700+ education and tech websites hijacked in huge ClickFix malware campaign

  • Under the engineering hood: Why Malwarebytes chose WordPress as its CMS

  • Think before you Click(Fix): Analyzing the ClickFix social engineering technique

  • Ghost CMS Vulnerability Exploited to Infect 700 Sites With ClickFix Malware