Huntress said on 7 October 2026 that since 10 September it has seen a phishing campaign that hosts its lure on Microsoft's real Power BI service so the first link gets past mail filters that trust the domain. Outlook phishing emails linked to a public report on app.powerbi.com showing a blurred form and a 'Download Reference' button. Huntress says attackers build these by publishing a dashboard from an account they control, often compromised or disposable, and sharing it publicly. No Power BI software flaw is involved. The button opened attacker pages (for example dailylifeproject[.]site and burnsworth[.]site) that fingerprinted the visitor (operating system, browser, screen size, automation signs, ISP and location), sent the details to a Telegram bot, allowed only Windows desktops and filtered out Microsoft or unknown ISPs, and redirected failures elsewhere. After a few seconds a script clicked a hidden link to download ScreenConnect.ClientSetup.exe from one attacker ScreenConnect tenant, with a different guest-access token for each lure. That first rogue client then installed a second ScreenConnect client tied to separate infrastructure (onthegotree[.]site), so removing one left access in place. In one incident a CMD file ran SCAutoFix.ps1 to install the second client and remove the first, the attackers ran a defence-evasion tool (HideUL_x64.exe), and a scheduled task, SCAutoRepairEvery2Min, re-ran the script every two minutes. Huntress saw a handful of endpoints hit by this delivery chain, and a retrospective hunt found the same unique ScreenConnect client and configuration on 22 more endpoints in separate incidents. Huntress says its SOC stopped the activity before it went further. Separate from the MSP360 and CSuite RMM phishing campaigns already on Cyberstack. (sources: Huntress; Cybersecurity News)
Huntress — Phishing campaign abuses Microsoft Power BI to deploy rogue RMMs (Sarah Reddish, 7 Oct 2026)
Searchlight Cyber researcher Kevin Gervot published a write-up on 7 October 2026 of two flaws in Discourse, the open-source forum platform Searchlight says powers more than 22,000 communities, including OpenAI, Atlassian and Cloudflare. Both were reported to Discourse on 9 June 2026 and affect default configurations. First, the color_scheme_id (and dark_scheme_id) cookie was written into a page tag without escaping and was not part of the anonymous page-cache key, so a single unauthenticated request could poison the cached page and run attacker JavaScript for every anonymous visitor, getting past Discourse's nonce-based Content Security Policy (CVE-2026-55674, CVSS 3.1 9.3). Second, a Trust Level 0 account, the lowest user level, could read files from the server through the profile picture upload. ImageMagick's policy blocked PostScript but still let Ghostscript handle files with a .pdf extension. Searchlight padded a JPEG so jpegoptim took about 25 seconds to process it, then used Ghostscript file operators that still work under its -dSAFER sandbox to swap the temporary file before ImageMagick computed the image's dominant colour. ImageMagick's MSVG coder then ran a Magick Scripting Language file that read a local file and saved it as an image, and Discourse handed it back as a normal upload URL (CVE-2026-55420). No Ghostscript zero-day was needed. Searchlight stopped at file read, but warns that reading Rails secrets such as secret_key_base could lead on to code execution. Discourse's CVE record for 55420 is titled remote code execution via PDF uploads, rates it CVSS 3.1 7.5 and says it needs certain non-default configurations. Discourse fixed its hosted sites on 15 June and published the ImageMagick fix on 30 June and the cache fix on 28 July. The fix adds a default-deny ImageMagick policy that blocks all delegates (no Ghostscript) and allowlists only the coders Discourse needs, and makes the colour-scheme cookies strict integers that are part of the cache key. This is separate from Hacktron's libheif research on Cyberstack. No exploitation has been reported.
Searchlight Cyber — A JPEG, a Race, and a Ghost: Breaking Discourse's Image Pipeline (Kevin Gervot, 7 Oct 2026)
AWS security bulletin 2026-128-AWS (published 7 October 2026, US Pacific time; early 8 October Perth) says missing authorisation checks in Amazon Athena engine version 3 request handling could have let an authenticated user read limited query metadata from other AWS accounts, namely AWS account identifiers and SQL statement text. AWS says query results, credentials and Amazon S3 data were not affected, that it fixed the issue on 1 September 2026, and that it has confirmed no customer metadata was accessed. No customer action is required. AWS credits Oren Yomtov of Act Security through coordinated disclosure. The AWS-issued CVE record rates it CVSS 3.1 7.7 (scope changed, high confidentiality impact). It is a cross-tenant flaw in a managed service, so there is nothing to patch on the customer side.
Microsoft told Microsoft 365 admins on 7 October 2026 (Message Center MC1490905) that Defender for Office 365 is extending Teams URL protection to links hidden inside QR codes. After a message is delivered, Defender extracts URLs from QR code images, checks them, and puts a warning on the affected message in both internal and external chats when a link is malicious. Organisations with Defender for Office 365 Plan 1 or Plan 2 and zero-hour auto purge (ZAP) for Teams turned on may also have eligible malicious internal messages blocked. URLs taken from QR codes will show up in the Advanced Hunting MessageUrlInfo table with UrlLocation set to QRCode. The feature is part of existing Teams URL protection and needs no end-user setup; general availability runs from early October to early November 2026. QR code phishing hides a malicious address in an image so text-based link scanning misses it. Sources: Microsoft 365 Message Center (public archive copy); Microsoft Learn; report: Cyber Security News 10 Oct.
MC1490905 — Defender for Office 365: post-delivery protection for malicious QR codes in Microsoft Teams (published 7 Oct 2026; Message Center archive copy)
A researcher writing as m00dy published findings on 7 October 2026, coordinated through CERT/CC (VU#765030), and CVE-2026-93034 was published on 8 October for SGLang, a widely used serving engine for large language models. SGLang's processes pass internal messages over ZeroMQ. By default they use Python Pickle (SGLANG_USE_PICKLE_IPC=true), which can run code while loading data; operators can switch to MessagePack, but the MessagePack path still accepts a PickleWrapper record and immediately unpickles it without a type allowlist or authentication. On a stock v0.5.18 service running a real model across two GPUs, an unauthenticated peer on a second machine got its code to run inside the detokenizer process, with the setting both on and off. In the default single-node layout the receiver is local to the host; it becomes reachable over the network when data-parallel attention is enabled with a non-loopback --dist-init-addr. Source inspection found the same code in v0.5.20 and main on 1 October; the researcher had not assessed v0.5.21 (released 2 October), so no fixed version is confirmed. CVSS 3.1 9.8 (CERT/CC). No exploitation reported. (source: CVE record)
m00dy — Disabling Pickle did not remove Pickle (SGLang, CVE-2026-93034 / VU#765030; 7 Oct 2026)
IBM published a security bulletin on 7 October 2026 for IBM Security Verify Access 10.0 to 10.0.9.2 and IBM Verify Identity Access 11.0 to 11.0.3, the access-management and reverse-proxy gateway (formerly ISAM/WebSEAL) many organisations put in front of web applications for single sign-on and multi-factor login. The worst issues, CVE-2026-78401 and CVE-2026-78406 (both CVSS 9.8), let a remote attacker with no login run arbitrary code through deserialisation of untrusted data (CWE-502). The bulletin lists 22 CVEs in all. It also covers CVE-2026-16916 (9.1, authenticated code execution through a protection mechanism failure in the appliance), CVE-2026-17189 (8.2, reflected cross-site scripting), CVE-2026-19878 (6.5, bypass of extra authentication workflow steps) and denial-of-service, log-forgery and local management interface flaws. Appliance and container editions are both affected. IBM lists no workarounds and no exploitation. (source: IBM)
IBM — Security Bulletin: vulnerabilities addressed in IBM Verify Identity Access and IBM Security Verify Access (7 Oct 2026)
GitGuardian reported on 7 October 2026 that GhostAction, the CI/CD supply-chain campaign it first exposed in September 2025, ran a new wave from 31 August 2026. Using what appear to be stolen GitHub credentials, the attacker committed a workflow named 'Github Actions Security' to 772 public repositories belonging to 373 GitHub users and organisations by 30 September; each workflow lists the repository's own secret names (scraped from earlier workflow history) and posts their values over plain HTTP to a bare IP address. Together the injected workflows target 2,577 secrets, most often SSH keys and Azure, AWS and database credentials. A variant named 'Security Check' (security-check.yml) appeared in seven repositories from 7 September and tags each theft with an injection ID. GitHub held most runs for approval: of 3,669 runs GitGuardian collected across 605 repositories, 499 executed in 32 repositories and 336 completed, exfiltrating 26 secrets from 13 repositories, with new runs still triggering. Only 124 repositories (16%) had been cleaned in public history by 5 October. In 92 cases the attacker simply re-pointed workflows left over from earlier waves, so GitGuardian says the campaign never really stopped. It also found 13 victim repositories used for GitHub Actions cryptomining by at least four other campaigns, including an XMRig miner hidden in the Docker image of the popular kuafuai/DevOpsGPT project, and thinks the stolen GitHub credentials are being traded among several actors. Researchers at Cynative independently spotted the commits. UPDATE (9 Oct): StepSecurity and Socket documented a new burst on 8 October in which two compromised maintainer accounts committed a workflow named 'security-audit.yml' ('Add security audit workflow') to every repository they could write to: 27 under kitao (Takashi Kitao, author of the 18,400-star pyxel game engine, published to PyPI and crates.io) from 13:20 UTC, then 318 under henrywoo (39 source repositories and 279 forks) plus Uber's uber/athenadriver in a 16-minute window from 21:10 UTC. This version sends named Actions secrets and also scans the working tree and full git history for cloud, AI and SaaS keys (AWS, Anthropic, OpenAI, OpenRouter, GitHub and GitLab tokens), posting them in cleartext to 193.32.204[.]199; StepSecurity confirmed from the athenadriver run log that exfiltration completed four seconds after the run started. Socket's 9 October update says more than 500 GitHub accounts have committed the workflow to tens of thousands of repositories since 7 October, including organisation-owned ones reached through compromised contributors. No malicious PyPI or crates.io releases had been published as of 9 October, but the stolen publishing tokens stay usable until rotated. UPDATE 13 October (GitGuardian, as also reported by The Hacker News on 15 Oct): GitGuardian says the campaign accelerated sharply between 8 and 12 October, peaking on 8–9 October. Its dataset now shows the actor targeted more than 30,000 repositories across 3,000 users and organisations by compromising the GitHub accounts of about 1,900 users, attempting to extract 3,174 secrets from public repositories' CI. The payload went first to 160 previously seen victims and, 20 minutes later, to about 28,000 new repositories on 8–9 October; later versions also capture the lines around each AWS key ID found in git history so the actor can pair it with its secret access key. Further waves on 10 and 12 October used the same payload but could not extract any secrets, because GitHub began preventing the malicious workflow from running on 9 October. Some victims of this burst had already been hit in the September 2025 and August 2026 waves, which GitGuardian reads as the actor relying on secrets that were never rotated. UPDATE 19 October 2026 (OpenSourceMalware, 11 and 16 Oct; The Hacker News, 19 Oct): the operator has moved beyond workflow injection. On 9 October it registered my-github[.]com, which resolves (with wildcard DNS) to the same collector, 193.32.204[.]199, and that day emailed an open-source maintainer a fake GitHub 'new sign-in from Moscow' security alert from noreply@my-github[.]com. The link opened a reverse-proxy phishing page that captured the password and authenticator code and ran injected JavaScript that silently approved a GitHub OAuth device flow with repo, workflow and read:org scopes, sending the resulting token to my-github[.]com on TCP 8443, so the victim never sees or types a device code. According to The Hacker News, a lookalike my-gitlab[.]com was registered 17 days before my-github[.]com and follows the same naming pattern, but there is no evidence it is run by the same actor; OpenSourceMalware lists it as an unconfirmed watchlist domain. Separately, OpenSourceMalware, starting from a find by Jamf's Thijs Xhaflaire, documented a second stage in 94 repositories belonging to 66 owners, most of them dormant for over a year: after a Stage 1 workflow on 8 October, the main burst on 15 October (UTC) committed a .vscode/tasks.json set to run on folder open, a fake font public/fonts/fa-solid-400.woff2 (the file name used by North Korea's PolinRider campaign) that is really a JavaScript stealer, and a second workflow under names such as lint.yml, ci-check.yml, build-check.yml, repo-health.yml, audit.yml, code-review.yml or test-suite.yml with more credential patterns and a secret-gist fallback. Opening the project in VS Code runs the stealer, which takes SSH keys, cloud credentials, npm and PyPI tokens, GitHub tokens, kubeconfig credentials, and any Anthropic, OpenAI or Gemini logins and keys, posts them over plain HTTP to port 8080 on the same server and installs a cron job (marked #ghc2) that fetches and runs server-supplied code every five minutes; deleting the font file does not remove the cron job. OpenSourceMalware says ten repositories also carry a workflow for self-hosted runners (such as cache-cleanup.yml) that asks the cloud metadata service for the machine's role credentials, reads SSH keys and credential files and installs the same cron job; The Hacker News says that workflow also reads credential files for Hugging Face and other tools. This stage does not need GitHub Actions to run, so GitHub's 9 October block does not stop it. OpenSourceMalware found no shared indicators with North Korean malware and calls it a copycat of PolinRider's technique. OpenSourceMalware has since updated its post: it now counts 1,230 repositories owned by 391 GitHub accounts carrying the fake font and hidden VS Code task. (sources: OpenSourceMalware; The Hacker News) Correction (29 Oct 2026): the 19 October update is now linked to OpenSourceMalware's two posts and The Hacker News report; the stealer's credential list now follows OpenSourceMalware, the my-gitlab[.]com detail follows The Hacker News, and an unlinked SecurityWeek credit was removed. (sources: OpenSourceMalware; The Hacker News)
CVE-2026-107181 was published on 7 October 2026 for a flaw in Telegram Desktop before 7.2.9 that researcher beaksec described in a technical write-up first published on 3 October and updated on 7 October. When a link is opened from outside Telegram (for example from a browser), a second instance passes it to the running app over a local inter-process channel that did not escape its record separator (CWE-143). A crafted tg:// link with an unescaped semicolon could therefore inject an extra command that reached an old internal helper, once used for release publishing, which uploads a local file to a chat without any permission check or prompt. Files that can be taken include Telegram's tdata session keys, so an attacker can take over the account when no local passcode is set. The demonstrated chain was on Windows with version 6.9.3 and is said to work through 7.2.8; it also relies on group files downloading automatically and on anyone being able to add the victim to a group, and the browser may ask before launching the app. Links clicked inside Telegram are not affected. Telegram fixed it in commit db3405699f (16 September) and released 7.2.9 on 17 September 2026 without a separate advisory; CVSS 3.1 is 8.1. No exploitation in the wild is known. (sources: CVE record; ThreatWire)
Let's Encrypt announced on 7 October 2026 that every certificate it issues or renews on or after 10 February 2027 will be valid for 64 days by default instead of 90, unless the subscriber picks one of the shorter 45-day or roughly six-day profiles it has already announced. The last 90-day certificates should expire on 11 May 2027, and existing certificates will not be revoked. The staging environment switches to 64-day certificates on 14 October 2026 so administrators can check renewal behaviour early. Clients that support ACME Renewal Information (ARI) will be told when to renew; anything renewing at a fixed number of days before expiry should move to about two-thirds of the lifetime (around day 43), and Let's Encrypt suggests searching cron jobs, wrapper scripts and runbooks for hard-coded values such as 83, 80 or 60. The authorisation reuse period also drops from 30 days to 10 days, and to seven hours in 2028, ahead of a 2029 industry limit; rate limits, ACME endpoints and issuance chains do not change. Default lifetimes are due to fall to 45 days in February 2028. Shorter lifetimes reduce the damage from a stolen key or a mis-issued certificate. (sources: Let's Encrypt; Cyber Security News)
Blockchain analytics firm Crystal Intelligence said on 6 October 2026 that it had analysed a leaked export of 5,692 chat messages from August 2025 to September 2026, mostly in Russian, which it says points to Silent Ransom Group (SRG), also tracked as Luna Moth, Chatty Spider, UNC3753 and Storm-0252, a group that emerged in 2022 after Conti shut down. Crystal says the group's internal deal board claims about US$207M paid by 27 firms between 3 April and 24 September 2026, and that 'On-chain evidence supports the scale but not the exact figures.' Crystal traced a collection wallet that received 344 BTC (about US$27M) in six weeks during April and May 2026 and about 2,675 BTC (around US$200M) over its lifetime, but says no single wallet can be tied to a specific victim payment. GuidePoint's research team (GRIT) said on 15 October that the internal ledger claims US$206.95 million from 27 organisations, with a median entry of US$6.0 million and a largest single entry of US$30 million, and that 'These are the actors' own internal records' and no victim has confirmed a payment. GRIT assesses with moderate-to-high confidence that the people in the chats operate under or alongside the SRG name, and says the targets are large law firms. The Record (8 October) reports that Chainalysis tied addresses in the leak to known SRG extortions but could not 'speak to the totality of claims', and The Register (9 October, Perth) says Crystal could not verify the US$207 million total. The group does not encrypt files: per Crystal, it sends fake invoices or calls staff posing as 'IT support', talks them into granting remote access, steals data and threatens to publish it. Correction (28 Oct 2026): the figures are now credited to Crystal Intelligence's and GuidePoint's own write-ups instead of Cyber Security News; removed the US$100,000 low figure, the nine-firm disclosure count and SRG's 8 October response, which only DataBreaches.net (HTTP 403 to Cyberstack) or Cyber Security News gave. (sources: Crystal Intelligence; GuidePoint GRIT; The Record; The Register)
Crystal Intelligence — Inside the Silent Ransom Group leak: how ransoms were laundered (6 Oct 2026)
On 7 October 2026 the Office of the Australian Information Commissioner announced a commissioner-initiated investigation into Shenzhen Qingcheng Future Technology Co. Ltd, maker of the HeyCyan companion app required for full use of Anko smart glasses sold by Kmart and other low-cost glasses sold via Big W Marketplace, Amazon and similar channels. After an Attorney-General request, the OAIC from 12 August 2026 sent preliminary inquiries to Meta, Kmart, Google and Shenzhen Qingcheng about recording, storage and access to personal information. Shenzhen Qingcheng (China) did not respond. Privacy Commissioner Carly Kind said that absence left no assurance the company protects Australians' personal information as required under APP 1, and that the investigation lets the OAIC use compulsory information-gathering powers. Kind has written to retailers selling HeyCyan-linked devices urging them to consider whether continued sale is appropriate. Opening an investigation is not a finding of breach. Parallel coverage: Kmart removed its $89 Anko camera glasses from online sale while still listing speaker-only glasses; ACS Information Age and SMH reported the retail pull alongside the OAIC action. (sources: OAIC media release; ACS Information Age)
OAIC — Opens investigation into maker of HeyCyan smart glasses phone app (7 Oct 2026)
German security firm Jakkaru disclosed on 7 October 2026 a business-logic flaw in iSolarCloud, the cloud platform Sungrow uses to manage solar plants, inverters and battery storage. One value of the login_type field in the (encrypted and signed) login request made the service authenticate the account named in the request and ignore the password, so anyone who knew a user's email address could log in, with no email or other login alert to the owner, and could then use account recovery to keep control. Customer and administrative accounts share the same management environment, so Jakkaru says a compromised administrator account could view and change plants, start or stop inverters and batteries, see registered organisations and users (including large German solar installers), and install custom firmware on any cloud-connected device. iSolarCloud runs four regional systems, European, Chinese, Australian and international, and the flaw was in all of them. Jakkaru says Sungrow published a hotfix within a day of being told and began a root-cause review. CVE-2026-107194 was published on 7 October, and no exploitation is reported. Sungrow is one of the largest inverter makers, with more than 1,000 GW of converters installed by Jakkaru's count, and is common on Australian rooftops. (sources: Jakkaru; Cyber Security News) Correction (22 Oct 2026): this card said no CVE had been published. CVE-2026-107194 was published on 7 October, and Sungrow's own advisory SGSA-202608-0028 (10 October) rates it CVSS 4.0 9.2 Critical, says the flaw was fully remediated on 25 August 2026 with no customer action needed, and describes the access an attacker gained as viewing the account's device list, operating status and historical monitoring data; the wider impact on this card is Jakkaru's assessment. (source: Sungrow security advisory SGSA-202608-0028)
Jakkaru — Sungrow vulnerability exposes gigawatts of power worldwide (7 Oct 2026)
