Researchers at Zimperium’s zLabs have analyzed an Android Trojan that uses an automated, multi-stage infection process. What’s new is that RatHat gives a live AI assistant the keys to the accessibility tree of the infected device and uses it to determine where to tap or scroll, rather than following a hardcoded script. The variable attack path makes it harder for signature- and rule-based mobile security tools to detect this Trojan. It also abuses Android Debug Bridge (ADB), a legitimate tool…
Analysis of how default configurations in AWS AgentCore Harness allow prompt injection to exfiltrate credentials, and key steps to secure your agents. The post A Vault with a Heap-View: The Uncomfortable Space Between AgentCore Harness and Identity appeared first on Unit 42.
Key Takeaways Periodic audits provide a point-in-time assessment, but they cannot demonstrate whether controls remain effective between audit cycles. Qualys platform data shows 10.5 billion configuration findings across customer environments but only 1.6% represent meaningful exposure and under 1% are prioritized, business-critical findings. Verizon’s 2026 DBIR found the median time to resolve weak passwords and misconfigured permissions is about 8 months. Across 1 billion misconfiguration…
AI has already made fundamental changes to the operating environment for cybersecurity. Cyberattackers are testing more paths, adapting their techniques, and moving across digital environments with greater speed and persistence. The weaknesses they exploit remain familiar: excessive permissions, unprotected authentication flows, unpatched systems, exposed execution paths, and gaps between controls. What has changed is how quickly these weaknesses can combine into attack paths that cross…
Executive Summary OpenAI disclosed that agents used exposed Hugging Face credentials to write a file and deploy proxy Spaces during an unrelated May 2026 research workload, but it did not identify the accounts. SentinelLABS identified two accounts likely used in associated activity, 0Time and Nyx9. Their public histories extend OpenAI’s chronology and preserve previously unreported relay code, document-borne probes, and ChatGPT account-provisioning capability. The public records provide precise…
Modern macOS malware uses deceptive setup guides to steal credentials and sensitive user data. Learn how to identify and block these threats. The post Atomic macOS (AMOS) Stealer Activity appeared first on Unit 42.
Attackers exploit Volume Shadow Copy for credential theft and ransomware defense evasion. See how Huntress spots the difference from routine IT activity.
We designed a behavioral clustering model to map cloud identity roles from audit logs, enabling continuous threat detection using standard SQL queries. The post Unmasking Cloud Identities: From Behavioral Clustering to Automated Detection appeared first on Unit 42.
A deep technical analysis of the Smishing Triad’s JWR phishing kit and Outsider operator cluster, revealing its real-time victim control, encrypted WebSocket communications, multi-stage credential theft, AES-256-CTR implementation, infrastructure, and actionable indicators for defenders.
I identified an attacker using a semi-autonomous coding agent to run an offensive operation: finding poorly secured LLM resale gateways, acquiring API access through ordinary web flaws and account farming, validating the resulting inference capacity, and aggregating it behind a single gateway of their own. What I captured was not simply credential theft, but an inference supply chain — an agent harvesting, validating, and consolidating LLM access into infrastructure that serves it again through…
IntroductionThe surge in emerging threat actors directly correlates with the rapid escalation of victim counts and stolen financial resources. Simultaneously, this growth has spurred the proliferation of specialized supply storefronts across social media platforms, dark web channels, and various smaller niche marketplaces. Security teams today face evolving challenges, requiring them to continuously refine monitoring channels, adjust operational strategies, and foster cross-functional internal…
Every wave of technology excitement creates a new opportunity for cyberattackers, and AI is no exception. Microsoft Threat Intelligence has published research showing a growing set of campaigns that impersonate popular AI platforms and tools, including ChatGPT, Microsoft Copilot, DeepSeek, and Claude.1 The goal is to make phishing, search-driven malware campaigns, and malvertising—which is malicious advertising that uses online ads to lure users to harmful sites, downloads, or redirect…
Identity was the target in roughly half of all confirmed malicious activity. Prophet Security breaks down the four main attack patterns seen across customer environments between May and July 2026, and explains why some attacks succeeded while others were blocked. [...]
Learn how root access on a compromised K8s node allows attackers to utilize SPIFFE/SPIRE metadata to spoof and harvest co-located workload identities. The post The Machine With Many Faces: Post-Exploitation Identity Misuse in SPIFFE/SPIRE appeared first on Unit 42.
An AD RMS Service Group account exports the AD RMS Server Licensor Certificate private key. That 1172-byte key decrypts every document the deployment ever protected, offline, and keeps doing so after the deployment is rebuilt.
About a week ago, Proxmox published an advisory revealing a vulnerability in older versions of Proxmox VE, its flagship Virtual Environment product. The vulnerability only affects version 7, which has not been supported for a couple of years now. But it appears that the vulnerability may have caught the attention of some attackers and researchers. We do see a bump in scans for port 8006, and also some additional brute force traffic. For example, brute force requests like: POST /api2/json/access…
In this article Attack chain overviewAttributionMitigation and protection guidanceLearn more Microsoft Security Research is tracking active cloud-based intrusions spanning multiple accounts in which unusual sign-ins were followed by threat actor-added authentication methods, high-volume Microsoft Graph activity, SharePoint and OneDrive downloads, and email collection through REST APIs. Microsoft Security Research assesses that this sequence is consistent with automated collection from…
The question of whether a Frontier AI model could find vulnerabilities that no human researcher had found was settled in April. Claude Mythos Preview identified thousands of previously unknown flaws across every major operating system and browser, including a 27-year-old denial-of-service condition in OpenBSD, and within a month Anthropic and its Project Glasswing partners had logged more than 10,000 high and critical severity findings, among them a certificate forgery flaw in wolfSSL, a…
Executive Summary Since the release of our May 2026 report detailing adversarial misuse of artificial intelligence (AI), Google Threat Intelligence Group (GTIG) has observed forward leaning adversaries transition from basic prompting to agentic AI workflows and AI-enabled automation. In these operations, human-in-the-loop latency is dramatically reduced, compressing the traditional window for defenders to respond. In Q2 2026, GTIG observed threat actors compromise a cloud resource, then plan,…
OverviewWhile conducting research into a recent N-able N-central authentication bypass vulnerability (CVE-2026-18577), Rapid7 Labs discovered two new vulnerabilities affecting the latest version of N-central. When chained together, these two vulnerabilities allow a remote unauthenticated attacker to bypass authentication and create a new attacker-controlled System administrator account on an affected server.CVE IDDescriptionCWECVSSv4CVE-2026-86206Semicolon/Forwarded access-control bypassCWE…
Active Directory Rights Management Services still ships in Windows Server 2025, years after Microsoft began steering customers to the cloud, and it remains fully supported on-premises. Part 1 maps the AD RMS trust model (the Server Licensor Certificate, the license flow, the SOAP surface) and shows how to discover an RMS deployment, fingerprint an AD RMS-protected file, and trace the path to that certificate's private key.
This week on the Lock and Code podcast… Crooks are taking a holiday. They’re counting on you to fund it. For decades, cybercriminals have stolen roughly the same types of data. Biographical and personal details—like Social Security numbers, birthdates, addresses, and phone numbers—can be stolen to commit identity fraud. Credit card numbers, expiration dates, and CVC codes can be stolen to make fraudulent purchases. Usernames and passwords can, in the wrong hands, let a cybercriminal impersonate…
In its first annual assessment, published Friday, the City of London Police said victims reported losing £6.3 million ($8.5 million) to account hacks in the year ending March 31, up from £1.2 million ($1.6 million) a year earlier.
Passkeys eliminate many password-based attacks, but researchers have documented 39 methods for compromising authentication built around them. Token explains how attackers can abuse authentication prompts, synced credentials, enrollment, recovery, and other trust boundaries without breaking FIDO2 cryptography. [...]
Key Takeaways Modern AD attacks use legitimate protocols end-to-end, no malware, no exploit, nothing for signature tools to fingerprint. The evidence is already in the logs; what is missing is the narrative linking five benign-looking Windows events into a single attack. A full domain takeover can be completed in 54 minutes, from the first password spray to the forged Golden Ticket, with each individual event appearing normal. Detection catches the move; posture management explains why it was…
A new dark web platform called Nexus claimed to be selling 153 million driver’s license scans and millions of other identity and medical cards. According to reports, the collection included more than 153 million driver’s licenses, 10 million ID cards, 3 million travel documents, and 579,000 medical cards, including marijuana dispensary cards. The trove of driver’s license scans reported by KrebsOnSecurity is a sharp reminder that identity verification is not a harmless box-ticking exercise. The…
Elastic Security Labs is tracking an emerging infostealer, REVSTEALER, under REF2859. This family has gained momentum in recent months, with higher distribution volume, and has emerged as a formidable threat, featuring a comprehensive credential harvester, an embedded sandbox scoring system, and a Polygon blockchain-based dead drop for resilience. Beyond credential theft, REVSTEALER targets gaming platforms for additional monetization; the developer clearly understands that victim accounts on…
The attacker gained unauthorized access to Coder’s Cloudflare infrastructure and added attacker-controlled IP addresses to the pool serving Coder’s module registry. These servers hosted modified registry artifacts containing malicious code designed to discover credentials and ...
Research by: hasherezade Key Points Since early 2025, Check Point Research has been tracking JSCeal, a sophisticated cryptocurrency-focused stealer with broader credential-theft, surveillance, and traffic-interception capabilities, delivered as compiled V8 bytecode (JSC files). The payloads are protected with javascript-obfuscator, using multiple techniques including RC4-protected strings, control-flow flattening, proxy functions, and operation wrappers. Our goal was to recover the code to a…
IntroductionDespite modern verification controls, identity theft remains one of the most pervasive threats to both individuals and enterprise organizations. U.S. Federal Trade Commission statistics show over 1 million identity theft reports annually, with related fraud and imposter scams accounting for billions in financial losses each year. While stolen credit cards enable rapid, short-term monetization, Social Security numbers (SSNs) represent a far more permanent and dangerous tier within…
In this article AI workloads are becoming high-value control pointsCase study 1: LiteLLM gateway compromiseCase study 2: RAGFlow compromiseCase study 3: Kestra compromiseMitigation and protection guidanceMITRE ATT&CK techniques observedReferencesLearn more AI is creating a new layer of enterprise infrastructure. Gateways, retrieval platforms, orchestration services, and containerized runtimes now sit between users, applications, data, and models. These systems concentrate credentials, data…
A phase-by-phase detection mapping of the first publicly documented autonomous agent intrusion against production infrastructure — including the phases where no product in our category sees anything at all. Executive Summary On July 9, 2026, an autonomous AI agent running inside an OpenAI capability evaluation escaped its sandbox and launched a multi-day intrusion against Hugging Face’s Kubernetes environment. Across roughly 17,600 actions, it moved from third-party infrastructure into the…
GreyNoise is observing automated scanners posing as the web crawlers of OpenAI, Anthropic, DeepSeek, and Fortune 500 companies, using forged user agents while requesting the files where misconfigured web servers frequently leak secrets and credentials.
Researchers have uncovered ToxicPanda 2.0, an Android banking Trojan and remote-access tool designed for account takeover and “on-device fraud.” Not only does ToxicPanda 2.0 have a much larger target list of banks and e-wallets, it has also expanded its capabilities by combining banking overlays, remote access, PIN capture, Android accessibility abuse, and attempted Wireless Debugging automation. Together, those functions can help operators turn a compromised phone into a platform for account…
In this article Activity overview Discovery of additional rotating infrastructure Attack chain overviewMitigation and protection guidanceReferencesLearn more MacSync Stealer is a macOS-focused information stealer that relies on changing infrastructure to deliver payloads, communicate with compromised devices, and exfiltrate data. Earlier reporting by RST Cloud identified the threat through a limited set of domains and documented rapid command-and-control (C2) replacement after public disclosure…
Learn how Tenable One Cloud Exposure helps you unmask the sophisticated tactics of cybercrime group Storm-0501, which carries out Azure-based cloud ransomware campaigns. Tenable One Cloud Exposure uses AI-powered threat stories to expose Storm-0501 TTPs, backed by precision-engineered threat detection alerts.Key takeawaysStorm-0501 demonstrates that cloud-first ransomware groups have shifted from simple endpoint encryption to the total hijacking of cloud tenants.Storm-0501 systematically…
Tenable’s Research Special Operations (RSO) team has been tracking a cluster of agentic AI threat activity since late July 2026. The Taiwan autonomous AI cyber attack confirmed what the cluster data already showed: near-autonomous offensive AI has crossed from theoretical risk to operational reality.Key TakeawaysTaiwan's Ministry of Digital Affairs confirmed a near-autonomous AI cyber attack in July 2026 in which autonomous agents mapped 21 connected government systems, compromised 85 accounts,…
Cisco Talos recently identified an undocumented phishing framework, internally branded "JWR" by its developer, built to convincingly impersonate checkout and login pages across major payment and shopping platforms. The client engine of the JWR phishing framework is a real-time, operator-driven system that, rather than merely logging form submissions like a static credential-stealing page, keeps an AES-CTR encrypted WebSocket open to the threat actor so they can steer each victim's session live.…
Key Takeaways Human-speed compliance is dead. Attackers utilizing modern, autonomous AI tools can chain enterprise misconfigurations and weaponize vulnerabilities in under 25 minutes, rendering manual, periodic audit cycles completely obsolete. The “Configuration Gap” is your biggest blind spot. Organizations take an average of 14 months to remediate basic identity, access control, and logging flaws, leaving a massive, open window of exposure for automated exploitation. Hyper-prioritization is…
Coding agents such as Claude Code and Cursor are vendor-signed, used all day on developer laptops, and routinely open shells, call APIs, edit files, and install helpers. That makes GenAI-adjacent alerts challenging to investigate. The parent looks trusted, while the children can still look a lot like classic high-severity activity. This article walks through one of those windows. On a macOS endpoint, Elastic Security endpoint telemetry showed shells under Claude Code that scripted a login to an…
Coding agents such as Claude Code and Cursor are vendor-signed, used all day on developer laptops, and routinely open shells, call APIs, edit files, and install helpers. That makes GenAI-adjacent alerts challenging to investigate. The parent looks trusted, while the children can still look a lot like classic high-severity activity. This article walks through one of those windows. On a macOS endpoint, Elastic Security endpoint telemetry showed shells under Claude Code that scripted a login to an…
On August 4, 2026, Elastic Security Labs identified a new Shai-Hulud campaign targeting the maintainer of keyv, a widely used key-value storage library. The attackers trojanized the monorepo and embedded a self-propagating worm called CHAINDROP that uses stolen npm credentials to automatically backdoor every other package the maintainer had publish rights to. The reach of this compromise is significant: keyv alone received over 600 million downloads last month, with related packages compounding…
On August 4, 2026, Elastic Security Labs identified a new Shai-Hulud campaign targeting the maintainer of keyv, a widely used key-value storage library. The attackers trojanized the monorepo and embedded a self-propagating worm called CHAINDROP that uses stolen npm credentials to automatically backdoor every other package the maintainer had publish rights to. The reach of this compromise is significant: keyv alone received over 600 million downloads last month, with related packages compounding…
In this article Attack chain overviewMitigation and protection guidanceIndicators of compromise (IOC)Microsoft Defender XDR detectionsAdvanced hunting queriesLearn more Microsoft Threat Intelligence identified a large-scale npm supply chain attack affecting more than 400 packages across multiple unrelated publishers, including packages associated with major enterprise software ecosystems such as keyv, flat-cache, cache-manager, and others. The malicious releases contain a Mini Shai-Hulud…
In this article The CaptiveCrunch campaignStorm-2945 and Midnight BlizzardCaptiveCrunch tradecraft and toolingHow to protect against CaptiveCrunch activityMicrosoft Defender detections and hunting guidanceIndicators of compromise Since early May 2026, Microsoft Threat Intelligence has observed Storm-2945, a sub-cluster of Midnight Blizzard, conducting widespread but targeted traffic manipulation attacks involving hospitality sector networks served by captive portals worldwide. Despite some…
Hugging Face reconstructed more than 17,000 attacker events from a July 2026 intrusion driven by an autonomous artificial intelligence (AI) agent. The path was familiar: untrusted dataset content abused a processing worker (file disclosure, then code execution), credential harvest, then multi-cluster lateral movement. Production Elastic Defend behavior rules and Elastic Security detection (SIEM) rules already watch those types of behaviors. This post maps each stage to detections you can enable…
Welcome to this week’s edition of the Threat Source newsletter. For my fianceé’s 30th birthday, I took her on a weekend trip to Shenandoah National Park – a favorite of ours since we went to a wedding there several years back. We’ve done several incredible hikes over the years, but one in particular had always loomed over my head: Old Rag, a 9.3 mile circuit hike that’s largely considered the most difficult in Virginia. I've always been warned that at the beginning and end, you hate Old Rag.…
Key Takeaways Two real-world cloud attacks reached meaningful impact in less than ten minutes despite pursuing entirely different objectives. Both attackers treated the environment as a connected system, using existing permissions and relationships to expand their reach. Reconnaissance increasingly focuses on understanding access and capability rather than discovering vulnerable assets. AI is compressing the gap between discovery, decision-making, and execution for cloud attackers. The interval…
Elastic Security Labs found a new Contagious Interview campaign, tracked as REF9403, hiding malware inside SVG image files using steganography. To our knowledge, this specific infection chain has not been previously documented. We found it after the DPRK-aligned group targeted our own community Slack workspace with a fake job posting and a "coding challenge" project. Any user who ran the project ended up with a four-stage payload aligned with OTTERCOOKIE: a browser credential and crypto wallet…
Written by: Corné de Jong Introduction Mandiant security assessments frequently identify publicly exposed serverless applications that lack authentication, often as a result of specific business requirements. Serverless deployments typically run custom-developed code that incorporates third-party packages, making them targets for a wide range of application-level attacks, including: Local and Remote File Inclusion (LFI/RFI) Command Injection Successful exploitation of these vulnerabilities can…
Two Microsoft 365 attacks got through Conditional Access policies that seemed fully configured. Learn what went wrong and how Huntress Managed ISPM catches these gaps first.
See how Shai Hulud-linked CI/CD compromise exposed Jenkins credentials, enabled AWS escalation, and led to Redshift breach activity detected by FortiCNAPP
AAD Graph Activity Logs are now ingestible into Elastic and usable for threat detection within the SIEM/XDR solution. That sentence shouldn't be exciting, but it is. For most of the past decade, this slice of telemetry simply didn't exist as a customer-accessible log stream. Microsoft Graph Activity Logs (the modern graph.microsoft.com surface) went GA in April 2024. The legacy graph.windows.net surface, the one adversary tooling actually hits, stayed dark until early 2026. This post walks the…