All news with #adversary in the middle tag

⚠️ Security researchers discovered two malicious Google Chrome extensions named Phantom Shuttle that intercept and exfiltrate credentials and session data from more than 170 targeted domains. After users pay for a subscription the add-ons enable a proxy 'smarty' mode, inject hard-coded proxy credentials, and route selected traffic through attacker-controlled proxies to establish a persistent Man‑in‑the‑Middle position. A recurring heartbeat to a command-and-control server forwards VIP emails, plaintext passwords and version details, enabling continuous monitoring and credential theft.

🔒 Two Chrome extensions in the Web Store, both published as Phantom Shuttle, are malicious plugins that hijack browser traffic and have been active since at least 2017, researchers report. Targeting users in China, the extensions pose as proxy and network-speed tools and prepend obfuscated code to the jQuery library to route requests through attacker-controlled proxies using hardcoded credentials and a PAC script. The plugins dynamically reconfigure Chrome proxy settings and route traffic for over 170 high-value domains, intercepting HTTP authentication challenges to capture form credentials, session cookies and API tokens while excluding local networks and the command-and-control domain to limit detection. At the time of reporting the extensions remained in Chrome's official marketplace; users are advised to install only extensions from reputable publishers and review requested permissions carefully.

📧 Sekoia.io researchers have identified a fresh wave of spear-phishing linked to the Russia-nexus intrusion set Star Blizzard (aka Calisto/ColdRiver) that targeted NGOs including Reporters Without Borders in May–June 2025. Operators impersonated trusted contacts via ProtonMail, using a custom Adversary-in-the-Middle kit to harvest credentials and relay 2FA prompts through compromised sites and redirectors. Observed tactics included a ZIP disguised as a .pdf, decoy encrypted PDFs instructing victims to open files in ProtonDrive, injected JavaScript to lock password-field focus, and an API-driven workflow for handling CAPTCHA and 2FA challenges, underscoring continued risk to Western organizations supporting Ukraine.

🔐 Researchers at Any.Run report a hybrid 2FA-phishing strain that fuses elements of Salty2FA and Tycoon2FA, producing payloads that evade detection rules tuned to either kit alone. The samples begin with Salty-style obfuscation and trampoline JavaScript, then shift into Tycoon’s DGA domains and AiTM execution chain. Analysts warn defenders to focus on behavioral patterns and fallback routines rather than static indicators of compromise.

📅 A targeted phishing campaign uses fake Calendly meeting invitations impersonating recruiters from major brands to harvest Google Workspace and Facebook Business credentials. The lures are professionally crafted—likely produced with AI—and direct victims through a CAPTCHA to an AiTM credential‑harvesting flow capable of bypassing some 2FA protections. Compromised ad manager accounts are then leveraged for malvertising, geo‑targeted attacks, device‑specific campaigns, or resale on illicit markets.

🔒 A 44-year-old man has been sentenced to seven years after pleading guilty to operating “evil twin” Wi‑Fi networks to harvest credentials and intimate images. AFP officers found a Wi‑Fi Pineapple, a laptop and a phone after airline staff reported a suspicious hotspot during a domestic flight. Forensic analysis recovered thousands of images and account credentials, and investigators linked malicious pages to airports and flights. Authorities advised users to disable automatic Wi‑Fi, use a reputable VPN, turn off file sharing and avoid sensitive transactions on public hotspots.

🛡️ A China-aligned group known as PlushDaemon has been observed deploying a previously undocumented network implant, codenamed EdgeStepper, to perform adversary-in-the-middle DNS attacks. ESET researchers found an ELF sample (internally called dns_cheat_v2) that forwards DNS traffic to attacker-controlled nodes, enabling update hijacking. Operators then deploy downloaders LittleDaemon and DaemonLogistics to install espionage backdoors.

🔒 ESET researchers disclosed a Go-based network backdoor dubbed EdgeStepper, used by the China-aligned actor PlushDaemon to reroute DNS queries and enable adversary-in-the-middle (AitM) attacks. EdgeStepper forces update-related DNS lookups to attacker-controlled nodes, delivering a malicious DLL that stages additional components. The chain targets update mechanisms for Chinese applications including Sogou Pinyin and ultimately fetches the SlowStepper backdoor to exfiltrate data.

🛡️ ESET researchers describe how the China-aligned actor PlushDaemon uses a previously undocumented network implant called EdgeStepper to perform adversary-in-the-middle hijacks of software update flows. EdgeStepper, a Go-based MIPS32 implant, redirects DNS traffic to malicious resolvers that reply with IPs of attacker-controlled hijacking nodes, causing legitimate updaters to fetch counterfeit components such as LittleDaemon. The analysis details the implant's AES-CBC encrypted configuration (notably using the GoFrame default key), iptables redirection of UDP/53 to a local port, and the downloader chain (LittleDaemon and DaemonicLogistics) that stages and deploys the SlowStepper backdoor on Windows hosts.

🔒 Hackers are exploiting LinkedIn direct messages to phish finance executives with messages claiming to invite recipients to an executive board and leading to credential-harvesting pages. Push Security says victims are redirected — including via a Google open redirect — to a Firebase-hosted 'LinkedIn Cloud Share' page that urges users to click a 'View with Microsoft' button. That flow then presents a Cloudflare Turnstile and a fake Microsoft sign-in used as an adversary-in-the-middle to capture credentials and session cookies; organizations should verify senders, avoid unsolicited links, and enforce MFA and conditional access.

🔒 ISO 15118-2-compliant EV charging implementations using the SLAC protocol are vulnerable to spoofed measurements that can enable man‑in‑the‑middle attacks between vehicles and chargers, tracked as CVE-2025-12357 (CVSS v4 7.2). The issue is an improper restriction of communication channel (CWE-923) and may be exploitable wirelessly at close range via electromagnetic induction. ISO recommends using TLS (required in ISO 15118-20) with certificate chaining; CISA advises minimizing network exposure, isolating control networks, and using secure remote access methods.

🔒 ConnectWise released a security update for Automate to fix two vulnerabilities including a critical 9.6-severity flaw (CVE-2025-11492) that can cause agents to use cleartext HTTP, enabling adversary-in-the-middle (AiTM) interception or modification of commands, credentials, and update payloads. A second 8.8-severity issue (CVE-2025-11493) omits integrity verification for update packages, allowing substituted malicious files. Cloud instances are patched to release 2025.9; on-premise administrators are urged to install the update within days.

🔒 The article warns that deploying synced passkeys introduces enterprise exposure because they inherit risks tied to cloud accounts and recovery processes. It highlights practical attack vectors — including AiTM-based authentication downgrades and malicious browser extensions — that can bypass or capture passkeys. The author recommends mandatory use of device-bound, hardware-backed authenticators and strict enrollment and recovery controls to preserve phishing-resistant access.

🔗 Phishing attacks are increasingly delivered outside traditional email — via social media, instant messaging, SMS, malvertising and in‑app messengers — making mail gateways insufficient. Attackers now send links from compromised accounts, targeted ads or SaaS messages and use fast‑rotating domains and advanced Attacker‑in‑the‑Middle (AitM) kits that obfuscate JavaScript and the DOM to evade network detection. Organizations often rely on user reports and URL blocking, but these approaches fail against rapid domain churn and client‑side stealth. Vendors such as Push Security propose browser‑level detection that monitors real‑time page behavior to identify AitM, session hijacking and credential theft.

🔒 VoidProxy is a newly observed phishing-as-a-service platform that leverages adversary-in-the-middle techniques to capture credentials, MFA codes, and session cookies from Microsoft 365 and Google accounts. Discovered by Okta Threat Intelligence, the service routes victims through shortened links and disposable domains protected by Cloudflare, serving CAPTCHAs and realistic login pages to selected targets. When credentials are entered, VoidProxy proxies requests to the real providers, records MFA responses, and extracts session cookies which are exposed in the platform admin panel for immediate abuse.

🔐 Okta has uncovered VoidProxy, a phishing-as-a-service operation that uses Adversary-in-the-Middle techniques to harvest Microsoft and Google credentials, MFA codes, and session tokens. The platform leverages compromised ESP accounts, URL shorteners, multiple redirects, Cloudflare Captcha and Cloudflare Workers to evade detection and hide infrastructure. Victims who enter credentials are proxied through an AitM server that captures session cookies and MFA responses, enabling account takeover. Okta recommends passkeys, security keys, device management, and session binding to mitigate the threat.