CTIPilot
AI-generated · no human review · verify critical claims against the linked source. how it works →
‹Tue · 19 May 2026›
All daily briefs →
Daily brief · UTC day

Tuesday, 19 May 2026

9 verified findings from 1 run · the settled record for this UTC day, in the classic brief order.

Criticality
Kind
Topic
Region
TL;DR · the day in one read
  1. 01MiniPlasma revived Windows Cloud Files flaw CVE-2020-17103 for a SYSTEM shell; Microsoft's June 2026 updates fix it. The researcher Chaotic Eclipse (also Nightmare Eclipse) released MiniPlasma on 2026-05-17, an exploit for the HsmOsBlockPlaceholderAccess routine of the Windows Cloud Files Mini Filter Driver (cldflt.sys), the flaw Google Project Zero reported in 2020 as CVE-2020-17103, which opened a SYSTEM command prompt from a standard user on Windows 11 with the May 2026 updates; BleepingComputer and Will Dormann reproduced it. Microsoft re-released CVE-2020-17103 on 2026-06-09 and recommends the June 2026 Windows updates to address MiniPlasma; it scores the flaw CVSS 3.1 7.0 and lists it as not exploited. →
  2. 02Grafana Labs: a GitHub workflow token missed after the TanStack npm compromise let attackers take its codebase. Grafana Labs confirmed on 2026-05-17 that an unauthorized party used a token for its GitHub environment to download its codebase, two days after the data-extortion group CoinbaseCartel listed it on its leak site, and refused the ransom demand. Its update of 2026-05-19 traces the breach to the TanStack npm supply-chain attack attributed to TeamPCP, detected on 2026-05-11, after which a missed GitHub workflow token let the attackers in; the download also took internal repositories holding business contact names and email addresses, while customer production systems and Grafana Cloud were not affected. →
  3. 03Symantec: fast16 corrupted LS-DYNA and AUTODYN simulations of uranium compression in nuclear detonations. Broadcom's Symantec and Carbon Black teams reported on 2026-05-16 that fast16, the circa-2005 sabotage framework SentinelOne first analyzed in April 2026, hooks LS-DYNA and AUTODYN through a boot-start filesystem driver and scales down simulation outputs only in high-explosive runs once the simulated material passes 30 g/cm³, a density uranium reaches only in an implosion device. It spreads only inside its own network and avoids hosts with endpoint-security products; Kim Zetter argues that, if it targeted Iran, it was contemporaneous with Stuxnet rather than a predecessor, while Broadcom still calls it pre-Stuxnet. →

01Active threats, incidents & disclosures6 items

NOTABLECVE-2026-46351 +2updatedNATOA2

BigBlueButton bbb-web before 3.0.21 and 3.0.23: predictable session tokens, an API checksum bypass and an SSRF (CVE-2026-46351, CVE-2026-46353, CVE-2026-46404)

BigBlueButton, an open-source virtual classroom, published three security advisories for its bbb-web component on 2026-05-13 (ENISA EUVD-2026-44996; BigBlueButton GHSA-7959-pf2v-xc4h, 2026-05-13; BigBlueButton GHSA-43hc-5g2m-cqff, 2026-05-13; BigBlueButton GHSA-xqm3-6q7q-4v5h, 2026-05-13). Before version 3.0.21, bbb-web generated conference session tokens with insufficiently secure randomness, so a user could determine the session tokens of other users in the same session and impersonate them (CVE-2026-46351) (BigBlueButton GHSA-7959-pf2v-xc4h, 2026-05-13; ENISA EUVD-2026-44996), and supplying a presentationUploadExternalUrl parameter bypassed checksum validation, so any user could send valid requests to some endpoints without a checksum (CVE-2026-46353) (BigBlueButton GHSA-43hc-5g2m-cqff, 2026-05-13; ENISA EUVD-2026-44998). Before 3.0.23, presentation URL validation did not restrict access to site-local and link-local addresses, and the fix also pins resolved IPs when following redirects; BigBlueButton says private-address assets may be compromised if the attacker has gained additional access (CVE-2026-46404) (BigBlueButton GHSA-xqm3-6q7q-4v5h, 2026-05-13; ENISA EUVD-2026-44997). The CVSS 3.1 scores in the CVE records are 8.1 for the first two, with low privileges and no user interaction, and 6.8 for the SSRF, with high privileges and a changed scope (ENISA EUVD-2026-44996; ENISA EUVD-2026-44998; ENISA EUVD-2026-44997). Germany's BSI summarized the set as letting a remote, authenticated attacker bypass security measures and disclose confidential information (translated from German) (BSI WID-SEC-2026-1568, 2026-05-18). No source reports exploitation.

Without the fix it is possible for a user to determine the sesionTokens of other users in the session.

Without this fix it is possible for any user to send valid requests to some endpoints without a checksum.

Private addresses assets may be compromised if the attacker has gained additional access.

BigBlueButton

A remote, authenticated attacker can exploit several vulnerabilities in BigBlueButton to bypass security measures and disclose confidential information.

BSI
Correctionrun 2026-10-21T1148Z-auditactionsaffected_productsbodyclassificationcvesevent_dateevidenceheadlinepriorityregionssectorssourcessourcing_notesummarytechniquestitle

BigBlueButton published the three advisories on 2026-05-13, not 2026-05-17, and BSI's advisory followed on 2026-05-18 (BigBlueButton GHSA-7959-pf2v-xc4h, 2026-05-13; BSI WID-SEC-2026-1568, 2026-05-18). No source names the research networks, Swiss cantonal schools or German education ministries said to run BigBlueButton; BSI describes it only as an open-source web-conferencing system (BSI WID-SEC-2026-1568, 2026-05-18). The checksum bypass lets any user send valid requests to some endpoints without a checksum, rather than arbitrary authenticated API calls (BigBlueButton GHSA-43hc-5g2m-cqff, 2026-05-13), and for the SSRF, which needs a high-privilege user, BigBlueButton says only that private-address assets may be compromised if the attacker has gained additional access (BigBlueButton GHSA-xqm3-6q7q-4v5h, 2026-05-13; ENISA EUVD-2026-44997).

threat19 May 05:00Zmulti-sourceOpen finding →
ROUTINEupdatedNATOB2

ARWINI, Lower Saxony's prescription audit body: police confirm a data outflow and name the Kairos group, which lists 2.87 TB for sale

According to the Hannover police directorate, data left the system of the Arbeitsgemeinschaft Wirtschaftlichkeitsprüfung Niedersachsen (ARWINI), which processes health and billing data of statutorily insured people in Lower Saxony and audits the cost-effectiveness of prescriptions on behalf of the statutory health insurers and the Kassenärztliche Vereinigung Niedersachsen (KVN), and the police confirmed to heise online that the Kairos ransomware group is behind the attack (Heise, 2026-05-18). The attack was discovered on 2026-05-04 and all systems were shut down at once, according to the KVN as relayed by Borns IT Blog (Borns IT Blog, 2026-05-16). ARWINI said that "based on current findings, it is likely that personal and also particularly sensitive data are affected" (translated from German), potentially contact and health data and billing information of statutorily insured patients, and that the extent of the outflow was not yet established (Deutsches Ärzteblatt, 2026-05-18); it put the worst case at up to 75,000 records, while its external data protection officer told heise online it was still unclear whether and which data had left (Heise, 2026-05-18). The KVN says the records it sends each quarter are pseudonymised, with patient data anonymised but doctor and practice numbers included (Heise, 2026-05-18), and HAZ and NDR estimated more than 70,000 records (Deutsches Ärzteblatt, 2026-05-18).

Kairos has listed a 2.87 TB dataset on its leak site since 2026-05-11 and threatens to sell it, with sample files that include letters from health insurers and doctors; as of 2026-05-18 the data had not been published, and whether the attackers hold that much is unverified (Heise, 2026-05-18). The Lower Saxony data protection commissioner has received a breach notification and is examining whether it came in time, and the police are in contact with foreign authorities over Kairos, among them Spanish investigators (Heise, 2026-05-18). The case matters as a statutory health-insurance audit body in the DACH region holding health and billing data; no source states how the attackers got in.

The ransomware group "Kairos" is behind the attack, as the police confirmed to heise online.

Arwini had stated that in the worst case up to 75,000 records could be affected.

Heise 2026-05-18

Based on current findings, it is likely that personal and also particularly sensitive data are affected

ARWINI (quoted by Deutsches Ärzteblatt)

A notification of a data breach has meanwhile also been received by the Lower Saxony data protection commissioner.

Heise 2026-05-18
Correctionrun 2026-10-21T1148Z-auditaffected_productsbodyclassificationevidenceheadlinepriorityregionssourcessourcing_notesummarytechniquestitle

The 70,000 figure is not ARWINI's: ARWINI said personal and particularly sensitive data were likely affected and put the worst case at up to 75,000 records, while HAZ and NDR estimated more than 70,000 (Deutsches Ärzteblatt, 2026-05-18; Heise, 2026-05-18). The KVN says the patient data it sends are anonymised, with doctor and practice numbers included, so the scope of patient data remains disputed (Heise, 2026-05-18). The authorities involved are the Hannover police directorate, which confirmed the outflow and named Kairos, and the Lower Saxony data protection commissioner, which received a breach notification and is examining whether it was made in time (Heise, 2026-05-18); no source reports a notification to the BSI or classifies ARWINI under KRITIS or NIS2 rules. The discovery date of 2026-05-04 comes from the KVN as relayed by Borns IT Blog (Borns IT Blog, 2026-05-16).

incident19 May 05:00Zmulti-sourceOpen finding →
NOTABLEupdatedNATOB2

Grafana Labs confirms GitHub breach claimed by CoinbaseCartel: a workflow token missed after the TanStack npm compromise let attackers download its code and internal repositories

Grafana Labs said in posts on X that an unauthorized party had obtained a token granting access to its GitHub environment and downloaded its codebase, that no customer data or personal information was accessed and customer systems were unaffected, and that it would not pay the attacker's extortion demand, following FBI guidance (The Hacker News, 2026-05-17; BleepingComputer, 2026-05-18). SecurityWeek dates the confirmation to Sunday 2026-05-17, two days after the group CoinbaseCartel listed Grafana on its leak site on 2026-05-15 (SecurityWeek, 2026-05-18). In its update of 2026-05-19, Grafana said the incident originated from the TanStack npm supply-chain attack via the Mini Shai-Hulud campaign, that it detected the malicious activity on 2026-05-11 and rotated a significant number of GitHub workflow tokens, but that a missed token, from a workflow it had first deemed not impacted, gave the attackers access to its repositories (Grafana Labs, 2026-05-19); The Hacker News attributes the TanStack attack to TeamPCP (The Hacker News, 2026-05-20). The download covered public and private source code and internal repositories some teams use for operational information, including business contact names and email addresses, but not production systems or the Grafana Cloud platform; the ransom demand from an unnamed actor arrived on 2026-05-16, and Grafana has since rotated automation tokens, implemented enhanced monitoring and audited all commits since the May 11 incident (Grafana Labs, 2026-05-19).

CoinbaseCartel, active since September 2025, steals data and extorts victims without encrypting files; SecurityWeek counts 105 victims on its site, while The Hacker News, citing Halcyon and FortiGuard Labs, puts the total at 170 (SecurityWeek, 2026-05-18; The Hacker News, 2026-05-17). Security firms link the group to ShinyHunters, Scattered Spider and Lapsus$ (SecurityWeek, 2026-05-18), but ShinyHunters told BleepingComputer that CoinbaseCartel is not linked to it (BleepingComputer, 2026-05-18). Grafana's paying customers include governments, banks and infrastructure operators, and more than 7,000 organizations use the product (BleepingComputer, 2026-05-18).

Grafana confirmed suffering a data breach on Sunday, two days after a cybercrime group listed the company on its leak website.

SecurityWeek 2026-05-18

We performed analysis and quickly rotated a significant number of GitHub workflow tokens, but a missed token led to the attackers gaining access to our GitHub repositories,

This includes business contact names and email addresses that would be exchanged in a professional relationship context, not information pulled from or processed through the use of production systems or the Grafana Cloud platform.

Grafana Labs (via The Hacker News)

After publishing this article, the ShinyHunters extortion gang told BleepingComputer that the CoinbaseCartel is not linked to their group or ransomware operation.

BleepingComputer 2026-05-18
Correctionrun 2026-10-21T1148Z-auditaffected_productsbodyclassificationentitiesevent_dateevidenceheadlineregionssectorssourcessourcing_notesummarytechniquestitle

No source on the May 2026 breach describes a pull_request_target or Pwn-Request mechanism: Grafana says the incident originated from the TanStack npm supply-chain attack via the Mini Shai-Hulud campaign, detected on 2026-05-11, and that a GitHub workflow token missed during the rotation that followed gave the attackers access (Grafana Labs, 2026-05-19), an attack The Hacker News attributes to TeamPCP (The Hacker News, 2026-05-20). The theft was not limited to source code: the downloaded content also included internal repositories holding operational information and business contact names and email addresses, while production systems and the Grafana Cloud platform were not involved (The Hacker News, 2026-05-20). The link between CoinbaseCartel and ShinyHunters is contested, since ShinyHunters told BleepingComputer the group is not linked to it (BleepingComputer, 2026-05-18), and SecurityWeek counts 105 victims on CoinbaseCartel's site rather than about 170 (SecurityWeek, 2026-05-18). No source places Grafana in Swiss or EU public-sector monitoring estates or calls for watching unofficial Grafana plugins; Grafana says it audited all commits for malicious activity (The Hacker News, 2026-05-20).

incident19 May 05:00Zmulti-sourceOpen finding →
NOTABLEupdatedNATOB2

CISA contractor's public GitHub repository exposed AWS GovCloud admin keys and plaintext passwords for internal CISA systems for up to six months

GitGuardian researcher Guillaume Valadon alerted KrebsOnSecurity on 2026-05-15 to a public GitHub repository named Private-CISA, kept by an employee of the government contractor Nightwing, that held cloud keys, tokens, plaintext passwords, logs and files describing how CISA builds, tests and deploys software (Krebs on Security, 2026-05-18). One file held administrative credentials to three AWS GovCloud servers, which Philippe Caturegli of Seralys validated as authenticating to three GovCloud accounts at a high privilege level; a CSV of browser-saved passwords listed plaintext credentials for dozens of internal CISA systems, among them LZ-DSO (Landing Zone DevSecOps), the agency's secure code development environment; and the archive also held plaintext credentials to CISA's internal Artifactory package repository (Krebs on Security, 2026-05-18). The commit logs show the administrator had disabled GitHub's default setting that blocks publishing SSH keys or other secrets, and many passwords were each platform's name followed by the current year (Krebs on Security, 2026-05-18).

The repository was created on 2025-11-13 and taken offline shortly after KrebsOnSecurity and Seralys notified CISA, but the AWS keys stayed valid for another 48 hours (Krebs on Security, 2026-05-18); Gizmodo puts the exposure at about six months, possibly much shorter depending on when each item was added (Gizmodo, 2026-05-19). Caturegli suspects the contractor used the repository to synchronize files between a work laptop and a home computer, and calls the Artifactory credentials a prime place to move laterally and backdoor software packages (Krebs on Security, 2026-05-18). CISA said it was investigating, that there was no indication any sensitive data was compromised, and that it was adding safeguards (Krebs on Security, 2026-05-18). For public bodies the case is a transferable pattern: a contractor's own repository holding an agency's privileged cloud and build-system credentials.

Valadon said the exposed CISA credentials represent a textbook example of poor security hygiene, noting that the commit logs in the offending GitHub account show that the CISA administrator disabled the default setting in GitHub that blocks users from publishing SSH keys or other secrets in public code repositories.

Caturegli said he validated that the exposed credentials could authenticate to three AWS GovCloud accounts at a high privilege level.

But Caturegli said the exposed AWS keys inexplicably continued to remain valid for another 48 hours.

Krebs on Security 2026-05-18
Correctionrun 2026-10-21T1148Z-auditaffected_productsbodyclassificationentitiesevent_dateevidenceheadlineregionssectorssourcessourcing_notesummarytechniquestitlepriority

LZ-DSO, which appears to stand for Landing Zone DevSecOps, the agency's secure code development environment, was one of the dozens of internal systems whose passwords sat in the exposed CSV; the plaintext credentials to CISA's internal Artifactory package repository were a separate item in the archive (Krebs on Security, 2026-05-18). CISA did not link the exposure to staffing: it said there was no indication that any sensitive data was compromised and that it was working to add safeguards, while the loss of nearly a third of its workforce is Krebs's own context (Krebs on Security, 2026-05-18). That the repository synchronized files between a work laptop and a home computer is Caturegli's suspicion, and his point about the Artifactory credentials is a risk he describes, a prime place to move laterally and backdoor software packages, rather than an established write access (Krebs on Security, 2026-05-18). The exposure ran from the repository's creation on 2025-11-13 until it was taken offline in mid-May 2026, up to about six months (Gizmodo, 2026-05-19).

incident19 May 05:00Zmulti-sourceOpen finding →
ROUTINEupdatedNATOA2

INTERPOL Operation Ramz: 13 MENA countries arrest 201 suspects, seize 53 servers and dismantle a phishing-as-a-service site in Algeria

INTERPOL's Operation Ramz ran from October 2025 to 2026-02-28 across Algeria, Bahrain, Egypt, Iraq, Jordan, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Tunisia and the UAE and focused on phishing and malware threats and on cyber scams; it led to 201 arrests, identified 382 further suspects and 3,867 victims, seized 53 servers and spread nearly 8,000 pieces of data and intelligence among the participants (INTERPOL, 2026-05-18; The Hacker News, 2026-05-18; Help Net Security, 2026-05-18). In Algeria authorities dismantled a phishing-as-a-service website, seizing a server, a computer, a mobile phone and hard drives with phishing software and scripts and taking one suspect into custody, and Moroccan authorities seized devices holding banking data and phishing software, with three people now in judicial proceedings (INTERPOL, 2026-05-18). In Qatar the operation identified compromised devices whose owners did not know they were being used to spread malicious threats, and in Oman investigators disabled a server in a private residence that held sensitive information and suffered from critical vulnerabilities, including a malware infection (INTERPOL, 2026-05-18). Jordanian police raided a fraud operation whose fake trading platform shut down once victims had deposited money and found that the 15 people running it were trafficking victims recruited from Asia with false job offers whose passports had been confiscated; two suspected organizers were arrested (INTERPOL, 2026-05-18). Group-IB, Kaspersky, the Shadowserver Foundation, Team Cymru and TrendAI supported the operation, which the EU and the Council of Europe partly funded through the CyberSouth+ project (INTERPOL, 2026-05-18).

A first-of-its-kind cybercrime operation in the MENA region has led to the arrest of 201 individuals, with a further 382 suspects identified.

In Algeria, a website offering phishing as a service was identified and dismantled as part of Operation Ramz.

A raid uncovered 15 individuals carrying out the scams, but investigators determined that they were victims of human trafficking who had been recruited under the false promise of employment from their home countries in Asia.

INTERPOL 2026-05-18
Correctionrun 2026-10-21T1148Z-auditaffected_productsbodyclassificationevidenceheadlineregionssectorssourcessourcing_notesummarytagstechniquestitlepriority

INTERPOL says the operation focused on phishing and malware threats and on cyber scams that inflict severe cost to the region; it names no targeting of EU banking customers or payment systems, and no source forecasts follow-on advisories or links the Jordan case to Europol work (INTERPOL, 2026-05-18). In Jordan police raided a computer used for a fake trading-platform fraud and found that the 15 people running it were trafficking victims recruited from Asia whose passports had been confiscated, and two suspected organizers were arrested (INTERPOL, 2026-05-18). The server in Oman sat in a private residence and held sensitive information its owner could legitimately access, but suffered from critical vulnerabilities including a malware infection and was disabled (INTERPOL, 2026-05-18). The Algerian takedown also put one suspect in custody, and in Qatar the operation found compromised devices being used, without their owners' knowledge, to spread malicious threats (INTERPOL, 2026-05-18).

threat19 May 05:00Zmulti-sourceOpen finding →
ROUTINEupdatedNATOB2

7-Eleven confirms breach of franchisee documents after ShinyHunters claims 600,000+ Salesforce records

7-Eleven's notification letters went out on 2026-05-01 (Cybersecurity Dive, 2026-05-20) and say an unauthorized third party gained access on 2026-04-08 to systems used to store franchisee documents (Security Affairs, 2026-05-18), which held personal information from the franchise application process, including names and addresses; the Vermont filing adds Social Security numbers and the Massachusetts filing Social Security numbers and driver's license data (Cybersecurity Dive, 2026-05-20). Filings in Massachusetts, Maine and Vermont count 47, 2 and 1 affected people, the letters offer up to 24 months of IDX identity-theft protection, and 7-Eleven says it has no reason to believe customer data was affected (Cybersecurity Dive, 2026-05-20); the notices became public on 2026-05-18 (SecurityWeek, 2026-05-18). ShinyHunters had listed 7-Eleven on its leak site on 2026-04-17 claiming more than 600,000 Salesforce records with personal and corporate data, demanded a ransom by 2026-04-21 and then offered the data for $250,000 on a hacker forum (SecurityWeek, 2026-05-18); 7-Eleven did not respond when asked about that claim (Cybersecurity Dive, 2026-05-20). Its relevance lies in the actor: SecurityWeek says ShinyHunters has targeted organizations' Salesforce instances since mid-2025 through phishing, abuse of third-party integrations or misconfigurations rather than Salesforce vulnerabilities, and that the group and affiliated actors recently claimed Instructure, Vimeo, Wynn Resorts, Vercel and Medtronic (SecurityWeek, 2026-05-18), while Security Affairs lists the European Commission among its earlier claims (Security Affairs, 2026-05-18). No source says how the 7-Eleven systems were accessed.

We recently discovered that on April 8, 2026, an unauthorized third party gained access to certain 7-Eleven systems used to store franchisee documents.

7-Eleven (letter quoted by Security Affairs)

The intrusions resulted from phishing, abuse of third-party integrations, or misconfigurations, rather than vulnerabilities in Salesforce products or systems.

SecurityWeek 2026-05-18

The breach impacted 47 people in Massachusetts, 2 in Maine and one in Vermont, according to each state’s filing.

Cybersecurity Dive 2026-05-20

We are aware of the situation, following our security protocols, and are in the process of alerting appropriate authorities. No sensitive personal information (PI) or customer proprietary network information (CPNI) data was exfiltrated by the threat actor.

Charter (via CyberInsider)
Updaterun 2026-05-25-d675ef38sectorssourcesbody

ShinyHunters added Charter Communications, which operates the Spectrum brand, to its leak site in the week of 2026-05-23, claiming over 42 million records containing personal information and threatening to leak them unless negotiations opened before 2026-05-27; the nature of the data had not been independently verified (CyberInsider, 2026-05-23). Charter confirmed a cybersecurity incident and told CyberInsider it was "following our security protocols" and alerting authorities, adding: "No sensitive personal information (PI) or customer proprietary network information (CPNI) data was exfiltrated by the threat actor." (CyberInsider, 2026-05-23). Charter did not disclose how the attackers gained access, and CyberInsider links the claim to the ongoing campaign against Salesforce environments (CyberInsider, 2026-05-23). Troy Hunt's weekly update of 2026-05-24 relays a post listing Charter among three new ShinyHunters claims (Troy Hunt, 2026-05-24).

Correctionrun 2026-10-21T1148Z-auditbodyclassificationentitiesevidenceheadlinepriorityregionssectorssourcessourcing_notesummarytechniquestitle

7-Eleven confirmed a breach of systems storing franchisee documents, not ShinyHunters' claim of 600,000+ Salesforce records: its notification letters date the access to 2026-04-08, the filings in Massachusetts, Maine and Vermont count 47, 2 and 1 affected people, and 7-Eleven did not respond when asked about the 600,000-record claim (Security Affairs, 2026-05-18; Cybersecurity Dive, 2026-05-20). SecurityWeek lists Instructure, Vimeo, Wynn Resorts, Vercel and Medtronic as recent claims by ShinyHunters and affiliated actors and describes the group's Salesforce intrusions since mid-2025 as phishing, third-party-integration abuse or misconfigurations; it does not place those victims in one campaign with 7-Eleven or name a vector for the 7-Eleven breach (SecurityWeek, 2026-05-18). Likewise, Charter did not disclose how the attackers got in, and CyberInsider only links the Charter claim to the Salesforce-focused campaign (CyberInsider, 2026-05-23). CyberInsider later updated its report: the attackers leaked the Charter dataset publicly, and Have I Been Pwned counts 4.9 million unique email addresses with names, phone numbers and physical addresses (CyberInsider, 2026-05-23).

incident19 May 05:00Zmulti-sourceOpen finding →
NOTABLECVE-2020-17103updatedNATOA1

MiniPlasma: Chaotic Eclipse PoC turns the 2020 Windows Cloud Files driver flaw CVE-2020-17103 into a SYSTEM shell on patched Windows 11; June 2026 updates fix it

Chaotic Eclipse, also known as Nightmare Eclipse, released source code and a compiled exploit for MiniPlasma on GitHub, saying that the flaw James Forshaw of Google Project Zero reported in September 2020 in the HsmOsBlockPlaceholderAccess routine of the Cloud Filter driver cldflt.sys, assigned CVE-2020-17103 and reportedly fixed in December 2020, was still present and that Google's original proof of concept worked unchanged (BleepingComputer, 2026-05-17). The exploit appears to abuse how the driver handles registry key creation through the undocumented CfAbortHydration API, which Forshaw's report said could create arbitrary registry keys in the .DEFAULT user hive without proper access checks (BleepingComputer, 2026-05-17). BleepingComputer, using a standard user account on a fully patched Windows 11 Pro system with the May 2026 updates, got a command prompt with SYSTEM privileges, and Will Dormann confirmed the exploit on the latest public Windows 11 but not on the latest Insider Preview Canary build (BleepingComputer, 2026-05-17); the researcher says success may vary because it is a race condition and that all Windows versions are likely affected (The Hacker News, 2026-05-18).

Microsoft re-released CVE-2020-17103 on 2026-06-09 to comprehensively address the flaw "recently publicly referred to as 'Mini-Plasma'" and recommends installing the June 2026 Windows updates; it rates it Important with a CVSS 3.1 base score of 7.0 and lists it as not exploited (Microsoft MSRC, CVE-2020-17103). MiniPlasma is the latest in a 2026 series from the same researcher after BlueHammer, RedSun and the UnDefend tool, all three of which were later seen exploited in attacks, and the YellowKey and GreenPlasma exploits (BleepingComputer, 2026-05-17).

BleepingComputer tested the exploit on a fully patched Windows 11 Pro system running the latest May 2026 Patch Tuesday updates.

Forshaw's original report said that the flaw could allow arbitrary registry keys to be created in the .DEFAULT user hive without proper access checks, potentially enabling privilege escalation.

BleepingComputer 2026-05-17

To comprehensively address the vulnerability identified by CVE-2020-17103 and recently publicly referred to as "Mini-Plasma" Microsoft recommends installing the June 2026 updates for your Windows operating systems.

Microsoft MSRC 2026-06-09
Correctionrun 2026-10-21T1148Z-auditaffected_productsbodyclassificationcvesevent_dateevidenceheadlinesectorssourcessourcing_notesummarytagstechniquestitle

Microsoft rates CVE-2020-17103 at CVSS 3.1 7.0, not 7.8, and re-released it on 2026-06-09, recommending the June 2026 Windows updates to comprehensively address the flaw now called Mini-Plasma, so it is no longer unpatched (Microsoft MSRC, CVE-2020-17103). The reproduction on fully patched Windows 11 Pro with the May 2026 updates was BleepingComputer's own test with a standard user account, and Will Dormann confirmed it separately on the latest public Windows 11 (BleepingComputer, 2026-05-17). MiniPlasma is not the third proof of concept in the series: BlueHammer, RedSun and UnDefend came first, followed by YellowKey and GreenPlasma (BleepingComputer, 2026-05-17). The monitoring the reporting gives is ThreatLocker's pair of registry paths; no source advises driver-load monitoring or says that disabling Cloud Files or OneDrive integration removes the attack surface (BleepingComputer, 2026-05-17).

vulnerability19 May 05:00Zmulti-sourceOpen finding →
ROUTINECVE-2026-42231 +4updatedNATOA2

n8n CVE-2026-42231 et al.: five authenticated prototype-pollution and node flaws (CVSS 4.0 9.4) let workflow editors reach code execution or read server files

n8n fixed five flaws, all scored CVSS 4.0 9.4 and all requiring an authenticated user with permission to create or modify workflows, in two batches. On 2026-04-22 it published CVE-2026-42231, prototype pollution through the xml2js parser of XML webhook bodies (n8n, GHSA-q5f4-99jv-pgg5, 2026-04-22), and CVE-2026-42232, prototype pollution through the XML node that leads to code execution when combined with other nodes (n8n, GHSA-hqr4-h3xv-9m3r, 2026-04-22), both fixed in 1.123.32, 2.17.4 and 2.18.1. On 2026-05-13 it published CVE-2026-44789, prototype pollution through an unvalidated pagination parameter in the HTTP Request node (n8n, GHSA-c8xv-5998-g76h, 2026-05-13); CVE-2026-44790, CLI-flag injection in the Git node's Push operation that reads arbitrary files from the server (n8n, GHSA-57g9-58c2-xjg3, 2026-05-13); and CVE-2026-44791, a bypass of the CVE-2026-42232 fix in the XML node (n8n, GHSA-wrwr-h859-xh2r, 2026-05-13), all three fixed in 1.123.43, 2.20.7 and 2.22.1. The Hacker News reported the set on 2026-05-18 (The Hacker News, 2026-05-18). No source reports exploitation.

Correctionrun 2026-10-11T1307Z-audittitleheadlinesummaryevent_dateprioritytechniquesaffected_productscvessourcesevidenceclassificationbodydeep_divedeep_dive_categoryregionssectors

n8n published the first two advisories on 2026-04-22 and the remaining three on 2026-05-13 (n8n, GHSA-q5f4-99jv-pgg5, 2026-04-22) (n8n, GHSA-c8xv-5998-g76h, 2026-05-13); a host patched only to 1.123.32, 2.17.4 or 2.18.1 is still exposed to the second batch. The advisories describe workarounds, restricting workflow editing and excluding the affected node through NODES_EXCLUDE (n8n, GHSA-57g9-58c2-xjg3, 2026-05-13). CVE-2026-44790 reads arbitrary files (n8n, GHSA-57g9-58c2-xjg3, 2026-05-13), and CVE-2026-44791 bypasses the XML node fix for CVE-2026-42232 (n8n, GHSA-wrwr-h859-xh2r, 2026-05-13). The priority moves from notable to routine: every flaw needs an authenticated user who can create or modify workflows, and no source reports exploitation.

vulnerability19 May 05:00Zmulti-sourceOpen finding →
Sources: n8n · The Hacker News

03Research, reports & policy1 item

NOTABLEupdatedNATOB1

Symantec and Carbon Black: the fast16 sabotage framework corrupted LS-DYNA and AUTODYN nuclear-detonation simulations; Kim Zetter argues it was contemporaneous with Stuxnet

Broadcom's Symantec and Carbon Black teams published on 2026-05-16 an analysis of fast16, the sabotage framework that SentinelOne first analyzed in April 2026 and whose oldest components date from around 2005, and confirm that its hook engine targets LS-DYNA and AUTODYN, applications used to simulate problems such as vehicle crashworthiness, material modelling and explosive simulation (Broadcom Security, 2026-05-16; The Hacker News, 2026-05-18). The framework combines a service binary that embeds a Lua 5.0 virtual machine, a boot-start filesystem filter driver that patches Intel-compiled executables on the fly as they are read from disk, and 101 byte-pattern rules in nine or ten groups for different builds of the two programs (Broadcom Security, 2026-05-16). Its hooks act only in high-explosive simulations and scale down outputs such as pressure or stress once the simulated material density passes 30 g/cm³, the threshold Broadcom says uranium reaches only under the shock compression of an implosion device, which it reads as sabotage of nuclear weapons research (Broadcom Security, 2026-05-16).

Before installing, fast16 checks for 18 endpoint-security registry keys and refuses to spread if any is present; it then registers itself as a service whose file takes the creation dates and permissions of a system binary, loads its driver as a SCSI-class filter at the next boot, and persists by hijacking a chosen application through the Image File Execution Options Debugger value (Broadcom Security, 2026-05-16). It spreads only inside its own network, enumerating domains, servers and shares and, for hosts in local address ranges, copying itself to their admin share under the logged-on user's credentials and creating a remote service (Broadcom Security, 2026-05-16). Kim Zetter, citing Symantec researchers and nuclear experts, argues that if fast16 targeted Iran in 2005 it was contemporaneous with Stuxnet rather than its predecessor, and that it fed false simulation data to engineers rather than altering weapons (Kim Zetter / ZERO DAY, 2026-05-16); Broadcom still calls it a pre-Stuxnet tool (Broadcom Security, 2026-05-16).

The malware checks for the density of the material being simulated and only acts when that value passes 30 g/cm³, the threshold uranium can only reach under the shock compression of an implosion device.

Broadcom Security 2026-05-16

Fast16 didn't predate Stuxnet but was contemporaneous with it. It also wasn't aimed at altering nuclear weapons but was simply feeding false data to engineers about the nuclear detonation tests they were conducting, in order to trick them into believing the tests were failing.

Kim Zetter / ZERO DAY 2026-05-16

Organizations concerned about similar threats capable of sabotage should iterate regularly across their endpoints to inventory loaded drivers, flagging any that are unsigned or unfamiliar.

Broadcom Security 2026-05-16
Correctionrun 2026-10-21T1148Z-auditaffected_productsbodyclassificationevent_dateevidenceheadlinesourcessourcing_notesummarytechniquestitle

SentinelOne's first public analysis of fast16 appeared in April 2026 and dated its oldest components to around 2005, about two years before Stuxnet first became active (Broadcom Security, 2026-05-16); no source places that disclosure at LABScon or reports earlier speculation that fast16 attacked centrifuges. The two framings are not settled: Kim Zetter argues that if fast16 targeted Iran in 2005 it was contemporaneous with Stuxnet rather than its predecessor (Kim Zetter / ZERO DAY, 2026-05-16), while Broadcom's analysis, published on 2026-05-16 rather than 2026-05-18, still calls it a pre-Stuxnet tool (Broadcom Security, 2026-05-16). LS-DYNA and AUTODYN are applications for simulating problems such as vehicle crashworthiness, material modelling and explosive simulation, not weapons-design codes as such, and Broadcom does not claim the hook engine is the first documented technique of its kind; the detection it supports rests on the install artifacts it documents and on inventorying unsigned or unfamiliar drivers (Broadcom Security, 2026-05-16).

research19 May 05:00Zmulti-sourceOpen finding →
Verification & coverage notes1 run

2026-05-19-2505c918 · Claude Opus 4.7 · 11 entries published

  • Items dropped: CVE-2026-41702 (VMware Fusion 25H2 macOS, TOCTOU SETUID race condition, CVSS 7.8, Broadcom VMSA-2026-0003 dated 2026-05-14): dropped from § 2, did not clear § 2 inclusion gates (no in-the-wild exploitation, not CISA KEV, not ENISA EUVD exploited=true, CVSS < 9.0, local-only attack vector on developer-workstation product, not pre-auth-RCE-on-edge-software).
  • Single-source items: none in this run.
  • Reduced-confidence items: ARWINI (S2) MULTI-SOURCE but no direct ARWINI press release retrieved; investigating-authority statement (Polizeidirektion Hannover) reported via Deutsches Ärzteblatt and Heise Security; ARWINI's own quote on Art. 9 data scope is via Borns IT Blog citing the ARWINI statement, not the statement page itself.
  • Contradictions: chalk-tempalte attacker-key descriptor: OX Security (blog post) describes the embedded key as a "public key"; The Hacker News (article), reporting on the same OX research, describes it as a "private key". Brief reports the attacker-controlled-key fact without taking a side on the key-type modifier; defenders cross-checking the brief should expect both descriptors in coverage.
  • Stalled sub-agents: none. All four Phase-1 sub-agents returned within wall-clock budget (S1 504s · S2 482s · S3 219s · S4 590s).
  • Verification: five iterations ran with model rotation (Opus / Sonnet / Opus / Sonnet / Opus); each iteration's findings were applied as remediations before the next spawn. Iteration 5 (final, Opus) returned NEEDS_FIXES with two residual findings (truth=1, editorial=1): F1 (CVE-2026-42232 component attribution), applied post-cap, the brief now describes it as "XML Node Prototype Pollution to RCE" matching the cited GHSA title; F2 (per-GHSA permalink specificity), applied post-cap, CVE Summary Table rows now link to each specific GHSA permalink instead of the advisories listing index. Per v2.50 cap-breach policy the brief publishes after the cap regardless; verification_residual_count = 2 reflects the iteration-5 verdict, not the post-cap state of the brief.
  • Models reported: main agent Claude Opus 4.7 (claude-opus-4-7); all four research sub-agents Claude Sonnet 4.6 (claude-sonnet-4-6).
  • Coverage gaps: bright-talk-dbir (403 on Verizon DBIR 2026 webinar URL); verizon-dbir-2026 (full PDF not yet available at run time, landing-page summary only); anssi-fr (no in-window advisories, newest 2026-05-13); bsi-de (only updates to already-covered advisories in window); cert-eu (RSS empty for in-window, newest item 2026-05-06); inside-it-ch (known 403, no bridge subcommand); heise-sec (TollBit-gated per-article URLs, used RSS summary + WebSearch corroboration); sophos-xops (rotation-priority empty, no in-window content); trendmicro-research (rotation-priority SPA shell with no RSS); databreaches-net (rotation-priority 403, Wayback empty); dark-reading (article 403 on Iran ATG fuel-tank campaign expansion; corroborated via WebSearch summaries but freshest CNN/Security-Magazine primary is dated outside the 36 h window so the item was dropped); sec-disclosures-edgar (no Item 1.05 8-K filings in window); ico-uk (no enforcement actions in window, newest action 2026-05-11 outside window).

Unmatched action items (migrated)

  • Patch self-hosted n8n now; apply the later train 1.123.43 / 2.20.7 / 2.22.1 which covers all five CVSS 9.4 CVEs in the cluster. The earlier patch (1.123.32 / 2.17.4 / 2.18.1) addresses only -42231 / -42232 and leaves the follow-on cluster (-44789 pagination prototype pollution, -44790 Git-node arbitrary file read, -44791 XML-node patch bypass) exposed. See § 2 entry and § 5 deep dive in this brief. Additional: enforce SSO+MFA on the editor role, restrict workflow create/modify to a small admin group, disable Git node if not required.
  • Audit every pull_request_target GitHub Actions workflow on agency / OSS-component repositories for write-scoped tokens reachable from external forked-PR code. Set permissions: read-all at workflow level; separate privileged steps into a second workflow_run-gated workflow that runs only on merged code; require CODEOWNERS approval before CI on external PRs. Grafana Labs (§ 4 UPDATE in this brief) is the second high-profile Pwn-Request loss this week.
  • Audit Checkmarx Jenkins AST plugin installations across CI/CD estate for any version installed during 2026-05-09 01:25 UTC → 2026-05-10 08:47 UTC; flag version 2026.5.09. Inventory checkmarx/ast-github-action and checkmarx/kics-github-action consumers; remediated builds are 2.0.13-848 / 2.0.13-847. CxSAST on-premise is unaffected; the trojanised surfaces are the Marketplace plugins and GitHub Actions. See § 4 TeamPCP/Shai-Hulud UPDATE in this brief.
  • Scan internal npm caches and lockfiles for packages by deadcode09284814 (chalk-tempalte, @deadcode09284814/axios-util, axois-utils, color-style-utils); inspect .vscode/tasks.json and ~/.claude/settings.json on developer endpoints for injected persistence hooks. Block egress from CI runners and developer workstations to *.lhr.life and other suspicious tunnel-provider domains used as commodity C2 channels.
  • Audit Salesforce Connected App OAuth grants and Event Monitoring across public-sector Salesforce tenants: particularly third-party AI / RPA SaaS integrations. Alert on bulk Report Export events and high-volume SOQL API calls; enforce IP-range / Trusted-IP session policies; consider Salesforce Shield field-level encryption for partner/supplier PII. The ShinyHunters Salesforce-targeting pattern hitting 7-Eleven, Instructure, Vimeo, Wynn Resorts, Vercel, Medtronic is identity-side, not Salesforce-product-side.
  • Hunt for MiniPlasma / Chaotic Eclipse Windows LPE PoC use via Sysmon EID 13 (RegistryEvent / SetValue) on the .DEFAULT user hive from non-SYSTEM processes; pivot on registry keys under \Registry\User\Software\Policies\Microsoft\CloudFiles\BlockedApps* and \Registry\User\.DEFAULT\Volatile Environment* (ThreatLocker hunt guidance). No vendor patch yet; mitigations are limited.
  • Inventory GKV / cantonal health-insurance and prescription-audit data-processor relationships as in-scope NIS2 / KRITIS critical suppliers; rehearse the 72-hour GDPR Art. 33 breach-notification clock starting from a third-party's detection event, not your own. ARWINI (§ 1 in this brief) follows the NMDL/IGJ Netherlands pattern from 2026-05-14.

Migrated from briefs/2026-05-19.md (v2).