All news with #data exfil via tools tag

EvilAI Campaign: Malware Masquerading as AI Tools Worldwide

🛡️ Security researchers at Trend Micro detail a global campaign called EvilAI that distributes malware disguised as AI-enhanced productivity tools and legitimate applications. Attackers employ professional-looking interfaces, valid code-signing certificates issued to short-lived companies, and covert encoding techniques such as Unicode homoglyphs to hide malicious payloads and evade detection. The stager-focused malware — linked to families tracked as BaoLoader and TamperedChef — performs reconnaissance, exfiltrates browser data, maintains AES-encrypted C2 channels, and stages systems for follow-on payloads. Targets span manufacturing, government, healthcare, technology, and retail across Europe, the Americas and AMEA.

UNC6040: Proactive Hardening for SaaS and Salesforce

🔒 Google Threat Intelligence Group (GTIG) tracks UNC6040, a financially motivated cluster that uses telephone-based social engineering to compromise SaaS environments, primarily targeting Salesforce. Operators trick users into authorizing malicious connected apps—often a fake Data Loader—to extract large datasets. The guidance prioritizes identity hardening, strict OAuth and API governance, device trust, and targeted logging and SIEM detections to identify rapid exfiltration and cross‑SaaS pivots.

Notion 3.0 Agents Expose Prompt-Injection Risk to Data

⚠️ Notion 3.0 introduces AI agents that, the author argues, create a dangerous attack surface. The vulnerability exploits Simon Willson’s lethal trifecta—access to private data, exposure to untrusted content, and the ability to communicate externally—by hiding executable instructions in a white-on-white PDF that instructs the model to collect and exfiltrate client data via a constructed URL. The post warns that current agentic systems cannot reliably distinguish trusted commands from malicious inputs and urges caution before deployment.

First Malicious MCP Server Found in NPM Postmark Package

🛡️ Cybersecurity researchers at Koi Security reported the first observed malicious Model Context Protocol (MCP) server embedded in an npm package, a trojanized copy of the postmark-mcp library. The malicious change, introduced in version 1.0.16 in September 2025 by developer "phanpak", added a one-line backdoor that BCCs every outgoing email to phan@giftshop[.]club. Users who installed the package should remove it immediately, rotate any potentially exposed credentials, and review email logs for unauthorized BCC activity.

Dutch teens arrested for alleged espionage near Europol

🔍 Two Dutch 17-year-olds allegedly used a WiFi sniffer to spy near Europol, Eurojust, and the Canadian embassy in The Hague. They were reportedly recruited over Telegram and arrested after a tip from the national intelligence service, the AIVD. Europol says its systems show no signs of compromise. The suspects will remain in custody for at least two weeks while investigators probe the case.

New COLDRIVER ClickFix Campaign Uses BAITSWITCH, SIMPLEFIX

🔍 Zscaler details a new COLDRIVER ClickFix campaign that deploys two lightweight families: BAITSWITCH, a DLL downloader, and SIMPLEFIX, a PowerShell backdoor. Victims are lured to execute a malicious DLL via a fake CAPTCHA; BAITSWITCH fetches SIMPLEFIX while presenting a Google Drive decoy. The chain stores encrypted payloads in the Windows Registry, uses a PowerShell stager, and clears the Run dialog to erase traces. Zscaler notes the campaign targets NGOs, human-rights defenders, think tanks, and exiles connected to Russia.

Postmark MCP Connector Compromised via Malicious NPM

🔒 A malicious npm package named postmark-mcp was discovered inserting a hidden Bcc that forwarded copies of transactional emails to an attacker-controlled server. Koi Security identified the backdoor in version 1.0.16 after its risk engine flagged suspicious behavior, noting the package had been trusted across many prior releases. With roughly 1,500 weekly downloads, the single-line injection enabled broad exfiltration of password resets, invoices, and internal correspondence before the package was removed; Koi urges immediate removal, credential rotation, and audits of all MCP connectors.

ArcaneDoor Targets Cisco ASA Firewalls in New Campaign

🔒 Cisco has linked a renewed campaign exploiting Cisco ASA 5500-X devices to the espionage-focused ArcaneDoor threat actor. The operation leveraged zero-day flaws, notably CVE-2025-20333 and CVE-2025-20362, to implant malware, modify ROMMON for persistence and evade detection by disabling logging and intercepting CLI commands. Observed compromises affected older ASA models lacking Secure Boot/Trust Anchor protections; Cisco and national authorities urge immediate remediation. Temporary mitigations include disabling SSL/TLS VPN web services and IKEv2 client services while applying vendor fixes and conducting forensics.

New macOS XCSSET Variant Targets Browsers and Clipboard

🛡️ Microsoft Threat Intelligence reported a new macOS malware variant of XCSSET that introduces browser-targeting changes, clipboard hijacking, and additional persistence mechanisms. The update uses run-only compiled AppleScripts, enhanced obfuscation and encryption, and expands data theft to include Firefox. New modules implement clipper behavior and LaunchDaemon- and Git-based persistence. Users should inspect Xcode projects and avoid pasting sensitive clipboard content.

Roblox executors: cheat tools that bring security risks

⚠️ Downloading third-party Roblox "executors" — tools that inject and run unauthorized scripts in games — can lead to account bans and serious security incidents. Malicious actors distribute fake or trojanised versions of popular tools such as Synapse X and Solara, sometimes bundling ransomware or backdoors. These installers may ask users to disable antivirus protections, which is a clear warning sign. Parents should steer children toward official features and avoid unverified downloads to keep accounts and devices safe.

Critical ForcedLeak Flaw Exposed in Salesforce AgentForce

⚠️ Researchers at Noma Security disclosed a critical 9.4-severity vulnerability called ForcedLeak that affected Salesforce's AI agent platform AgentForce. The chain used indirect prompt injection via Web-to-Lead form fields to hide malicious instructions within CRM data, enabling potential theft of contact records and pipeline details. Salesforce has patched the issue by enforcing Trusted URLs and reclaiming an expired domain used in the attack proof-of-concept. Organizations are advised to apply updates, audit lead data for suspicious entries, and strengthen real-time prompt-injection detection and tool-calling guardrails.

Malicious Rust crates on Crates.io exfiltrate crypto keys

🔒Two malicious Rust crates published to Crates.io scanned developer systems at runtime to harvest cryptocurrency private keys and other secrets. The packages, faster_log and async_println, mimicked a legitimate logging crate to avoid detection and contained a hidden payload that searched files and environment variables for Ethereum-style hex keys, Solana-style Base58 strings, and bracketed byte arrays. Discovered by Socket, both crates were removed and the publisher accounts suspended; affected developers are advised to clean systems and move assets to new wallets.

XCSSET Evolves: New Clipboard, Firefox, Persistence Modules

🔍 Microsoft Threat Intelligence describes a new XCSSET variant that infects Xcode projects and expands capabilities to include clipboard hijacking, Firefox data theft, and additional persistence via LaunchDaemon entries. The actor uses run-only compiled AppleScripts, AES-based encryption, and layered obfuscation to evade analysis. A bnk submodule monitors and can replace wallet addresses in the clipboard while a new Mach-O binary targets Firefox data. Organizations are advised to patch promptly, inspect Xcode project sources, and deploy Microsoft Defender for Endpoint.

Chinese Backdoor Grants Year-Long Access to US Firms

🔐 Chinese state-linked actors deployed a custom Linux/BSD backdoor called BRICKSTORM on network edge appliances to maintain persistent access into U.S. legal, technology, SaaS and outsourcing firms. These implants averaged 393 days of undetected dwell time and were used to pivot to VMware vCenter/ESXi hosts, Windows systems, and Microsoft 365 mailboxes. Mandiant and Google TAG attribute the activity to UNC5221 and have released a scanner and hunting guidance to locate affected appliances.

Malicious npm Package Uses QR Code to Steal Cookies

🔍 A malicious npm package named Fezbox was discovered using QR-code steganography to conceal and deliver a credential-stealing payload. The package fetched a QR image from a remote URL, waited roughly 120 seconds, decoded embedded code and executed it to extract usernames and passwords from browser cookies. Socket's AI-based scanner flagged the behavior; the package, which had at least 327 downloads, was removed after a takedown request to the npm security team.

Google: Brickstorm malware stole data from U.S. orgs

🔒 Google researchers warn that the Go-based Brickstorm backdoor was used in prolonged espionage against U.S. technology, legal, SaaS, and BPO organizations, averaging a 393-day dwell time. Suspected activity from the UNC5221 cluster involved deploying the malware on appliances lacking EDR protection such as VMware vCenter/ESXi, where it acted as a web server, SOCKS proxy, file dropper, and remote shell. Operators used techniques like a malicious Java Servlet Filter (Bricksteal), VM cloning, and startup-script modifications to capture credentials and move laterally, then tunneled to exfiltrate emails via Microsoft Entra ID Enterprise Apps. Mandiant published a scanner and YARA rules to aid detection but cautions it may not catch all variants or persistence.

Brickstorm: Long-term Go-based Backdoor Targets US Orgs

🔒 Google researchers report suspected China-linked operators used a Go-based backdoor named Brickstorm to persistently exfiltrate data from U.S. technology, legal, SaaS and BPO organizations, with an average dwell time of 393 days. Brickstorm operated as a web server, file dropper, SOCKS relay and remote command executor while masquerading traffic as legitimate cloud services and targeting edge appliances that often lack EDR. GTIG attributes the activity to UNC5221, a cluster linked to Ivanti zero-day exploitation and custom tools like Spawnant and Zipline. Mandiant published a scanner with YARA rules but cautioned it may not detect all variants or persistence mechanisms.

Feds Tie Scattered Spider Duo to $115M in Ransoms

🔒 U.S. prosecutors have charged 19‑year‑old Thalha Jubair as a core member of Scattered Spider, alleging the group extorted at least $115 million from victims. Jubair and an alleged co‑conspirator, 18‑year‑old Owen Flowers, appeared in London court facing accusations tied to high‑profile attacks on retailers, public transit and U.S. healthcare providers. Authorities say the complaint links Jubair to a network of SIM‑swapping, SMS phishing and ransomware operations and to cryptocurrency servers used to launder proceeds.

QR Codes Used to Hide JavaScript Backdoor in npm Package

🔒 A malicious npm package called fezbox was discovered using layered obfuscation and QR-code steganography to conceal credential-stealing logic. Disguised as a benign JavaScript/TypeScript utility, importing the library triggered retrieval and execution of code hidden inside a remote QR image; the payload reads document.cookie and attempts to extract username and password pairs for exfiltration. Socket researchers highlighted a development-environment guard and a 120-second delay as anti-analysis measures; the package has been removed from GitHub and marked malicious.

Nimbus Manticore Intensifies Cyber-Espionage in Europe

🔍 Check Point Research reports that Iranian-linked actor Nimbus Manticore has escalated cyber-espionage operations across Western Europe, with heightened targeting of organizations in Denmark, Sweden and Portugal. Attackers impersonate recruiters and use convincing fake career portals to deliver personalized credentials and malicious archives. The campaign leverages evolved backdoors—first seen as Minibike, now observed as MiniJunk and MiniBrowse—and employs multi-stage DLL sideloading into legitimate Windows binaries, including Microsoft Defender components, alongside valid code-signing certificates and compiler-level obfuscation to evade detection. Infrastructure hosted via Azure App Service and shielded by Cloudflare provides redundancy and rapid command-and-control recovery.