VulnHub

Real-time Cybersecurity News

Critical

Thousands of malicious AI skills found capable of stealing data, running malware

Help Net Security
Actively Exploited
Malware

Overview

Researchers at ESET have identified over 25,000 potentially malicious AI skills among nearly 900,000 analyzed. These skills, which enable AI agents to perform various tasks online, can be exploited by attackers to steal sensitive data, execute malware, or alter the behavior of the AI systems. This discovery raises significant concerns about the security of AI tools, as they could be manipulated to harm users or systems. The findings highlight the need for vigilance in monitoring AI skills and ensuring that they are secure, as the misuse of these capabilities can lead to serious data breaches and other cybersecurity incidents.

Key Takeaways

  • Active Exploitation: This vulnerability is being actively exploited by attackers. Immediate action is recommended.
  • Affected Systems: AI skills and agents
  • Action Required: Users should regularly review and verify the AI skills they enable, ensuring they come from trusted sources.
  • Timeline: Newly disclosed

Original Article Summary

AI agents can browse the web, use external tools, execute commands, and perform tasks on behalf of users. Many rely on skills that define how they interact with services and data. Malicious skills can abuse those capabilities to steal data, execute malware, or manipulate an agent’s behavior, according to the H1 2026 ESET Threat Report. Malicious AI skills expand the attack surface An analysis of nearly 900,000 AI skills identified more than 25,000 suspicious skills … More → The post Thousands of malicious AI skills found capable of stealing data, running malware appeared first on Help Net Security.

Impact

AI skills and agents

Exploitation Status

This vulnerability is confirmed to be actively exploited by attackers in real-world attacks. Organizations should prioritize patching or implementing workarounds immediately.

Timeline

Newly disclosed

Remediation

Users should regularly review and verify the AI skills they enable, ensuring they come from trusted sources. Organizations may also need to implement stricter controls on AI skill deployment and usage.

Additional Information

This threat intelligence is aggregated from trusted cybersecurity sources. For the most up-to-date information, technical details, and official vendor guidance, please refer to the original article linked below.

Related Topics: This incident relates to Malware.

Read Original Article

Related Coverage

Ring adopts new TAKE encryption standard for smart home devices

SCM feed for Latest

Ring has adopted a new encryption standard called TAKE for its smart home devices. This standard uses a rotating set of encryption keys that are temporarily stored in the cloud, allowing Ring to secure active user features. The implementation of TAKE is designed to enhance the security of user data and improve the overall safety of smart home devices. While this development aims to bolster encryption practices, it raises questions about the security of cloud-stored keys and how they are managed. Users of Ring devices should stay informed about these changes to understand how their data is protected and what potential vulnerabilities may exist in the cloud storage approach.

Aug 26, 2026

Nimbus Manticore expands infrastructure and malware arsenal

SCM feed for Latest

Nimbus Manticore, linked to the Tortoiseshell hacking group, has expanded its capabilities by deploying a new SSH-based tunneling tool and a C++ backdoor that resembles its existing malware known as TWOSTROKE. This development indicates a shift in tactics, allowing attackers to establish more secure communications with compromised systems. The increase in their malware arsenal raises concerns for organizations that may be targeted, as it suggests a growing sophistication in their operations. Companies need to be vigilant and enhance their defenses against potential intrusions, especially those using SSH protocols. The implications of this escalation could lead to more successful breaches and data exfiltration if not addressed promptly.

Aug 26, 2026

Carhartt data breach claims inflated by synthetic data, analysis finds

SCM feed for Latest

On August 13, the hacking group ShinyHunters claimed to have leaked 50GB of data from Carhartt after demanding a ransom of $3.3 million. However, recent analysis suggests that the data leak may not be as significant as initially reported, with claims of inflated figures due to synthetic data. This incident raises concerns about the reliability of data breaches reported by cybercriminals and highlights the risks companies face from extortion attempts. If the data claims are exaggerated, it could lead to unnecessary panic among Carhartt's customers and stakeholders, as well as impact the company's reputation. It's crucial for organizations to remain vigilant and verify claims made by hackers to protect their interests.

Aug 26, 2026

Cyber threats nudge Trump to sign executive order on foreign equipment in U.S. energy infrastructure

CyberScoop

In response to ongoing cyber threats, President Trump signed an executive order aimed at protecting U.S. energy infrastructure from foreign-produced equipment that could pose national security risks. The order specifically prohibits the purchase and installation of such equipment, reflecting growing concerns over potential vulnerabilities in critical systems. This move targets equipment from foreign entities that may not adhere to U.S. security standards, highlighting the administration's focus on safeguarding essential infrastructure. The implications of this order could affect various energy projects and suppliers, especially those reliant on foreign technology. By tightening regulations on foreign equipment, the government aims to bolster national security and mitigate risks associated with cyber attacks.

Aug 26, 2026

Officials disrupt Chinese espionage operation that hit multiple federal agencies

CyberScoop

Authorities recently disrupted a Chinese espionage operation that had been infiltrating sensitive networks of multiple federal agencies for over eight years. The attackers, reportedly funded by the Chinese government, utilized a sophisticated hacking suite that enabled them to operate undetected for an extended period. The operation's exposure raises significant concerns about the security of federal systems and the potential for sensitive data breaches. This incident underscores the ongoing challenges that government agencies face in protecting their networks from foreign cyber threats. As investigations continue, the full scope of the damage and the specific agencies affected may become clearer.

Aug 26, 2026

Red Flags That Expose Fake North Korean IT Workers

darkreading

Researchers have identified that North Korean operatives are increasingly posing as IT workers to infiltrate organizations. These operatives have been refining their tactics, making them harder to detect. However, experts have pointed out several red flags that can help companies identify these fake IT workers before any harm is done. This situation is concerning as it highlights the potential for significant data breaches and cyberattacks against businesses that may unknowingly hire these individuals. Companies need to be vigilant and ensure they have robust vetting processes in place to protect themselves from these sophisticated threats.

Aug 26, 2026