Ghostcommit attack hides malicious AI instructions in images


Ghostcommit is a proof of concept that shows how AI assistants used to review software code can be tricked by hidden instructions embedded in images.

The academic ASSET Research Group showed that an attacker can place instructions inside an image file, point to it in an AGENTS.md file, and get an AI coding agent to follow those instructions during a later task.

A pull request is basically a formal “please review and add my changes” request that a developer sends before changes are added to the main version of a software project. Human reviewers and, increasingly, AI coding tools may review the changes before they’re accepted.

Read more…
Source:  MalwareBytes Labs


Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox


Related:

  • Florida ransomware negotiator convicted for helping ransomware gang extort US companies

    July 10, 2026

    Florida man Angelo Martino has been sentenced to more than five years in prison for conspiring with hackers to deploy ransomware during his job as a ransomware negotiator for a U.S. cybersecurity company. The U.S. Department of Justice confirmed the sentence on Thursday, noting that the government seized more than $10 million worth of cryptocurrency and assets. Martino ...

  • Accenture confirms breach after hacker steals 35GB of source code and other data

    July 9, 2026

    Accenture has confirmed suffering a cyberattack, days after threat actors started selling an archive allegedly coming from the firm. “We are aware of this isolated matter, and we have remediated its source. There is no impact to Accenture operations and service delivery,” Accenture said in a statement. It follows a relatively unknown threat actor called 888 posting ...

  • GigaWiper: Anatomy of a destructive backdoor assembled from multiple malware

    July 9, 2026

    In October 2025, Microsoft Threat Intelligence identified destructive wiping activity and uncovered a sophisticated Go programming language (Golang)-based backdoor we now track as GigaWiper, a versatile implant that combines robust command-and-control (C2) capabilities with multiple destructive payloads, including disk wiping, fake ransomware, and system-level sabotage. GigaWiper is particularly notable for its makeup. It’s not a single, ...

  • 6.9 million driver’s license numbers stolen from AssuranceAmerica

    July 9, 2026

    Insurance provider AssuranceAmerica has confirmed a data breach affecting the personal information and driver’s license numbers of up to 6.9 million people. AssuranceAmerica provides car and rental insurance to customers across 14 US states through a network of over 9,500 independent agents. The breach notice letter also mentions information about customers’ auto insurance policies and accounts, their drivers and ...

  • Vidar Stealer Unmasked: Code Signing Abuse, Go Loaders and File Inflation

    July 7, 2026

    In April 2026, Unit 42 researchers identified a financially motivated campaign delivering Vidar stealer and the XMRig cryptocurrency miner to consumer and small- and medium-sized business victims worldwide. Attackers lure victims via malvertising to pages for downloading files that impersonate cracked versions of copyright-protected software. Upon execution, the loader drops and runs both Vidar stealer and ...

  • Fake Netflix, Coca-Cola, and FIFA job scams target marketers

    July 7, 2026

    Attackers are impersonating major companies and recruiters to target marketing professionals, using trusted services and browser tricks to make the scam look legitimate. A BleepingComputer article detailing the campaign found at least 34 domains impersonating high-value companies, including Netflix, Coca-Cola, Adidas, and FIFA. The lure is a fake job interview or scheduling request from a “recruiter” representing one of ...