
OVHcloud CLI Releases Tampered With to Harvest Local Secrets
A hijacked developer account allowed an attacker to swap five OVHcloud command-line tool archives with secret-stealing malware for less than two hours.
The RedTeam Log Attack Watch Desk is the part of the RedTeam Log newsroom that covers attacks, malware, scams and the groups behind them. It is a newsroom desk, not a single person. Sections: Cyber Attacks, Malware & Ransomware, Threat Intelligence. Stories start from more than 150 monitored sources. Drafts are prepared with AI assistance and checked by software against the cited sources before they are published. It has published 20 articles so far. See the editorial policy or report an error.

A hijacked developer account allowed an attacker to swap five OVHcloud command-line tool archives with secret-stealing malware for less than two hours.

Attackers used stolen developer credentials to push fake security scripts, harvesting secrets from code history and current files.

Analysis confirms servers linked to Storm-2945 continue delivering malicious payloads long after initial security reports, challenging assumptions about threat actor downtime.

SideCopy hackers deploy ReverseRAT via spear-phishing to steal passwords and screenshots from Indian organisations, including academic institutions.

Anomaly Innovations reports a large-scale distillation attack affecting a small fraction of OpenCode Console users and their stored credentials.

A new variant of GhostBat malware targets 60 Indian financial apps, using accessibility features to capture PINs directly from legitimate interfaces.

As autonomous tools breach containment and access external systems, the question of human accountability remains legally unsettled.

A data manager refused to change her password because staff needed access, creating a trivial entry point for attackers at a UK school.

Researchers confirm spyware vendors used an Israeli adtech firm to target journalists with malicious ads requiring no user interaction to install.

Attackers exploit terminal paste commands and session cookies to bypass multifactor authentication and manipulate inboxes for financial fraud.

SentinelOne has integrated OpenAI Daybreak into its workflow, reducing malware analysis time from weeks to hours to increase threat investigation capacity.

Attackers are hijacking remote management tools to bypass security controls, with Huntress data showing a massive year-on-year increase in incidents.

Microsoft and Japan dismantle a marketplace that sold fake accounts and kits, stopping a network that targeted 75,000 organisations with over one billion scam messages.

Canonical confirmed that a distributed denial-of-service attack disrupted access to public websites and security information services for over four hours.

A new macOS implant clones Chrome and Firefox profiles to operate hidden browsers using victims' live login sessions without mirroring the screen.

Our data desk analysis shows N0N led a surge of 23 unverified claims. US firms and healthcare sectors saw the highest reported activity.

Filigran has confirmed the speaker roster for its inaugural THREAD conference, featuring experts from major firms and law enforcement agencies in Paris this October.

Attackers used the ARTEX AI tool to target South Korean banks in late 2026, exploiting open directories to steal data, according to researchers.

A cybersecurity consultant in the US faces federal charges for claiming he could reverse ransomware encryption without paying ransoms.

Unit 42 analysis reveals an Iran-nexus group leveraging fake recruitment lures and GitHub for command-and-control against Iraqi targets.