North Korea's state hackers are no longer content to type prompts into public chatbots. One of the country's main espionage groups has begun running artificial intelligence (AI) offline on its own servers, connecting document-search tools to files in its possession, and collecting the software parts needed to build AI into its malware. South Korean security firm Genians says it uncovered the The incident highlights how adversaries continue to evolve their tradecraft, combining increasingly accessible tooling with targeted social engineering to slip past traditional perimeter defenses. Finally, maintain offline, tested backups and a clear communication plan so that business continuity decisions are made ahead of time rather than under pressure.
LexisNexis took its Diligence, Metabase API, and Newsdesk services offline as part of its response to unusual activity on servers hosted and managed by an unnamed third-party vendor. This development is consistent with broader industry trends, where threat actors increasingly reuse proven techniques and commodity tooling rather than investing in novel malware. Security teams should review their detection rules, keep threat-intelligence feeds current, and validate that incident-response runbooks are tested before an incident occurs.
Attacker-controlled instructions can make Atlassian's Rovo assistant collect Jira or Confluence data that a signed-in user can access, then send it to an outside server. Two security firms found that behavior independently, by different routes. Only one of those routes is confirmed closed. PromptArmor, an AI security firm, hid the instructions in content Rovo reads. It said an uploaded file was enough to make the assistant gather internal data and send it out through a URL request, with no separate approval step. The firm published on August 5, 2026 and said the chain still worked with Rovo's web-search option switched off. That bypass is single-sourced, and the report establishes the finding's status only on that date; a later remediation is not confirmed here. Varonis Threat Labs put the instructions in a link instead. It found that the rovoChatPrompt URL parameter would preload attacker instructions into Rovo Chat, so one click from an authenticated user was enough for Rovo to run them with that user's privileges and send the results to an attacker-controlled server.
WordPress has fixed a pre-authentication reflected cross-site scripting (XSS) flaw in its login screen that affects every version of the content management system. pwn.ai demonstrated how the flaw can be chained into PHP code execution on the server when a logged-in administrator interacts with an attacker-controlled page. Tracked as CVE-2026-64638 (CVSS score: 8.9), the high-severity The flaw has been cataloged as CVE-CVE-2026-64638, and technical details are now widely available, which means exploitation attempts often follow disclosure very quickly. Security teams should review their detection rules, keep threat-intelligence feeds current, and validate that incident-response runbooks are tested before an incident occurs.
WordPress has fixed a pre-authentication reflected cross-site scripting (XSS) flaw in its login screen that affects every version of the content management system. pwn.ai demonstrated how the flaw can be chained into PHP code execution on the server when a logged-in administrator interacts with an attacker-controlled page. Tracked as CVE-2026-64638 (CVSS score: 8.9), the high-severity vulnerability requires no attacker privileges. According to pwn.ai, which discovered the flaw and shared technical details with The Hacker News, the login-page XSS requires no authentication. Once a crafted username reaches the failed-login error page, the resulting JavaScript executes in the visitor's browser with no further interaction required on that page. The code-execution path requires a victim already logged in as an Administrator and explicit interaction with an attacker-controlled page. In pwn.ai's demonstration, that interaction is one ordinary click.
Cybersecurity researchers have called attention to an active "widespread email-driven phishing campaign" that employs adversary-in-the-middle (AitM) techniques to take control of Microsoft 365 accounts with an aim to identify key personnel involved in financial workflows and gather related email. "The campaign uses residential proxies to disguise malicious sign-ins as ordinary consumer traffic," Arctic Wolf Labs said. "Automated activity maintains compromised sessions at approximately eight-hour intervals." The activity is assessed to impact organizations across healthcare, education, manufacturing, government, and professional services sectors located in the U.S., Canada, and Europe. It shares tactical overlaps with Payroll Pirate attacks tracked by Microsoft under the moniker Storm-2755. Payroll Pirates is the designation assigned to a broader financially motivated threat cluster that involves hijacking the accounts of employees to reroute salary payments to attacker-controlled accounts. Some aspects of these campaigns have been documented since early 2025, with Microsoft tracking a related threat as Storm-2657.
PortSwigger says HTTP Terminator, an artificial intelligence (AI)-assisted research system built by James Kettle, generated and proved new HTTP desynchronization techniques after exploring 30,000 candidate desync vectors. PortSwigger said a separate human-guided discovery cascade also exposed a zero-day in Apache Traffic Server. Kettle said HTTP Terminator tested 30,000 websites where scanning was authorized through bug bounty or vulnerability disclosure programs and found roughly 700 vulnerable targets before deeper validation and RQP research. Kettle said those findings involved banks, government infrastructure, security products, and an airport. The research produced new desync triggers, a dual-matching Content-Length pattern, and a "dangling-byte" technique designed to make response queue poisoning (RQP) more reliable. RQP can potentially make a front end lose track of which back-end response belongs to which user, potentially exposing another user's response, including session cookies or API keys.
A GitHub issue opened by an account with no repository privileges was enough to execute code on the CI runners behind Anthropic's and Google's own coding-agent repositories. On OpenAI's, it was enough to hijack the next agent run. Novee Security ran the attack against each vendor's agent in the configuration that the vendor ships by default, and presented the work at Black Hat USA on August 5. Two CVEs came out of it. Both are patched. Gemini CLI carries the worst of the two. CVE-2026-12537 (CVSS 4 score: 10.0) is an OS command injection in the container launcher, reached through a crafted .gemini/.env file, which lets an unprivileged attacker run code on the host of a headless CI platform before the sandbox starts. It is fixed in Gemini CLI 0.39.1 and run-gemini-cli 0.1.22. In Claude Code, CVE-2026-54316 turned Hugging Face's public download counter into an exfiltration channel that leaked an API key one character at a time, and is fixed in 2.1.163. Every Claude Code release from 0.2.54 up to 2.1.163 is affected. Anthropic says exploitation required getting untrusted content into a Claude Code context.
Coinspect has identified CryptoJS.lib.WordArray.random() as the weak random number generator behind the Ill Bloom wallet drains. Introduced in the JavaScript cryptography library 12 years ago, the function supplied weak entropy that affected wallet apps used to generate recovery phrases. Coinspect's on-chain analysis puts the measured theft across two sweeps since late May at a lower bound of roughly $5.7 million. The blockchain security firm, which coined the Ill Bloom name in July, has now confirmed five applications that used the generator as an entropy source for recovery-phrase generation: Coinspect told The Hacker News that these are the same five wallets it referenced without naming in July. The firm said the vendor behind the browser extension mentioned in July told it that the extension had been discontinued and was never widely used. Coinspect cautioned that it cannot confirm it identified every vulnerable wallet. Other affected mobile or browser-extension wallets may also have existed but could not be examined because they had already been removed from app stores or extension marketplaces, or because vendors had replaced them with patched releases and the older vulnerable versions were no longer available.
Attackers broke into an organization's Oracle database through a SQL injection flaw in a public-facing web application, then installed a post-exploitation toolkit without writing an executable to disk. They fed Java source code to the database, let Oracle compile it into stored schema objects, and ran commands from inside the database engine. Huntress, which tracks the toolkit as khunt, The incident highlights how adversaries continue to evolve their tradecraft, combining increasingly accessible tooling with targeted social engineering to slip past traditional perimeter defenses. Finally, maintain offline, tested backups and a clear communication plan so that business continuity decisions are made ahead of time rather than under pressure.
Confirm this action.
Leaving now will discard your changes.