On this page
On this page
- 0. TL;DR
- 1. Active Threats, Trending Actors, Notable Incidents & Disclosures
- 2. Trending Vulnerabilities
- 3. Research & Investigative Reporting
- 4. Updates to Prior Coverage
- 5. Deep Dive — Linux cgroups v1 release_agent container escape (CVE-2022-0492) re-enters active exploitation
- 6. Action Items
- 7. Verification Notes
Tags (21)
Regions (5)
References (24)
- CVE-2024-21182
- CVE-2025-48595
- CVE-2022-0492
- CVE-2020-1472 ×2
- CVE-2025-8088 ×2
- CVE-2026-40402
- Gamaredon GammaPhish/GammaWorm — NTFS-ADS USB+network worm (Sekoia)
- NCSC-CH pre-event cyber advisory for the G7 Évian summit (DDoS/intel-collection/mobile targeting)
- Dashlane TOTP brute-force — encrypted vaults of <20 personal-plan users downloaded
- Attacker-built AI-orchestrated EDR-evasion testing lab (Sophos X-Ops)
- Sophos 2026 Active Adversary Report — identity-dominant root causes; Impacket/AnyDesk
- SVG phishing wave using application/ecmascript MIME to evade WAF/email pattern-matching (SANS ISC)
- Operation XENOFISCAL — SideCopy/APT36 XenoRAT via mshta/HTA vs Afghan provincial treasuries
- BleepingComputer
- CISA Known Exploited Vulnerabilities Catalog
- Help Net Security
- NCSC Switzerland — Im Fokus
- Oracle Critical Patch Updates
- SANS Internet Storm Center
- Security Affairs
- Sekoia.io blog
- Sophos X-Ops (incl. former Secureworks CTU)
- Palo Alto Networks Unit 42
- Seqrite Labs (Quick Heal Technologies research arm)
0. TL;DR
- Oracle WebLogic CVE-2024-21182 (CVSS 7.5) added to CISA KEV on evidence of active exploitation — an unauthenticated attacker reaching the T3 or IIOP listeners (default ports 7001/7002) gains unauthorized access to WebLogic-accessible data. Patched in Oracle's July 2024 CPU; the in-window signal is the fresh exploitation, not the 23-month-old fix. WebLogic remains common middleware in EU finance and public-sector estates (The Hacker News, 2026-06-02).
- Google patches an actively-exploited, High-severity Android zero-day, CVE-2025-48595, in the June 2026 bulletin — an Android Framework integer overflow giving no-interaction local privilege escalation across Android 14/15/16; Google reports "limited, targeted exploitation" (a profile consistent with commercial-spyware use, though no source attributes this case). Full fix requires the 2026-06-05 patch level (Android Security Bulletin, 2026-06-01).
- A four-year-old Linux container-escape, CVE-2022-0492, re-enters CISA KEV — the cgroup-v1
release_agentmissing-CAP_SYS_ADMINcheck lets a process in a permissively-profiled container execute code at host level. Today's deep dive (§ 5) covers the escape path and the mandatory-access-control hardening that closes it (CISA, 2026-06-02). - NCSC Switzerland issues a pre-event cyber advisory ahead of the G7 Évian summit (15–17 June) — the NCSC explicitly anticipates hacktivist DDoS against Swiss organisations (NCSC Switzerland, 2026-06-01); an independent threat map additionally flags state intelligence collection against hotel/telecom infrastructure and mobile-device targeting, echoing the NoName057(16) DDoS waves seen during Bürgenstock 2024 (ZENDATA, 2026-05-03). Most delegations transit Swiss infrastructure (Geneva–Vaud corridor).
- Dashlane discloses a TOTP brute-force that downloaded the encrypted vaults of fewer than 20 personal-plan users — attackers exhausted the bounded six-digit TOTP keyspace to register a new trusted device, the same new-device-registration kill chain as the 2022 LastPass breach. Vaults stay master-password-encrypted but face offline cracking (TechCrunch, 2026-06-02).
1. Active Threats, Trending Actors, Notable Incidents & Disclosures
NCSC Switzerland warns of cyber operations around the G7 Évian summit (15–17 June)
On 2026-06-01 Switzerland's National Cyber Security Centre published a pre-event advisory warning that the G7 summit in Évian (France, 15–17 June) is a high-value target and that it "expects disruptive maneuvers in cyberspace again" (NCSC Switzerland, 2026-06-01). Although the summit sits on French soil, most delegations transit Geneva Airport and lodge on the Swiss side (Geneva, Vaud, Valais), putting Swiss federal and cantonal administrations, conference-linked suppliers, and Swiss telecom operators in the blast radius. An independently published threat map for the event frames the expected activity against the template of the 2024 Bürgenstock summit, when the pro-Russia hacktivist collective NoName057(16) ran DDoS waves against Swiss federal sites and conference-linked organisations on each summit day; the same map additionally flags state intelligence collection against hotel and telecom infrastructure, rogue-base-station cellular interception, and social-engineering against event staff as plausible vectors (ZENDATA Cybersecurity, 2026-05-03). The NCSC advisory itself recommends generic protective measures and DDoS preparedness for organisations linked to the event.
Why it matters to us: Organisations operating in the Geneva–Vaud corridor and Swiss federal/cantonal SOCs should pre-stage DDoS mitigation playbooks now, review MFA on customer-facing identity providers, rotate administrative credentials before the event window, and brief travelling staff on mobile-device physical security; hunt for anomalous authentication spikes from the summit region and unexpected reattachment events in MDM/MDM-adjacent telemetry around 15–17 June.
Dashlane discloses TOTP brute-force that downloaded encrypted vaults of fewer than 20 users
Dashlane disclosed (on 2026-06-01, for an attack dated 2026-05-31) that an external actor brute-forced its TOTP second factor to download the encrypted vaults of fewer than 20 personal-plan accounts (TechCrunch, 2026-06-02). The technique abuses the bounded TOTP keyspace — one million six-digit codes per 30-second window — by submitting a high volume of attempts against the new-device-registration endpoint, where a single correct code registers a new trusted device that can then pull the vault (The Hacker News, 2026-06-02). Dashlane's rate-limiting locked the targeted accounts (since restored) and the company states its infrastructure was not compromised; vault contents remain encrypted under the user's master password, which Dashlane does not store, but weak master passwords now face offline cracking (BleepingComputer, 2026-06-01). This is structurally the same new-device-registration kill chain that enabled vault theft in the 2022 LastPass breach.
2. Trending Vulnerabilities
CVE-2024-21182 — Oracle WebLogic Server: unauthenticated T3/IIOP data access, KEV-listed on active exploitation
CISA added CVE-2024-21182 to the Known Exploited Vulnerabilities catalog on 2026-06-01 "based on evidence of active exploitation" (The Hacker News, 2026-06-02). The flaw (CVSS 7.5) lets an unauthenticated, network-positioned attacker abuse the T3 or IIOP protocol listeners — exposed by default on ports 7001/7002 — to obtain unauthorized access to WebLogic-accessible data, and on some configurations a more complete server compromise. It affects Oracle WebLogic Server 12.2.1.4.0 and 14.1.1.0.0 and was fixed in Oracle's July 2024 Critical Patch Update (Oracle CPU, 2024-07-16). The operationally relevant fact is the fresh exploitation against a patch that has been available for 23 months, not the FCEB remediation date attached to the KEV entry; WebLogic is heavily deployed J2EE middleware in EU financial-services and public-sector estates (Security Affairs, 2026-06-02). Defenders: apply the July 2024 (or later) CPU; block T3/IIOP at the perimeter and restrict it to internal admin subnets via WebLogic connection filters; alert on unauthenticated T3/IIOP initiators reaching 7001/7002 from external sources.
Changes since first coverage(1 prior appearance)
- 2026-06-022026-06-02
CVE-2025-48595 — Android Framework: actively-exploited integer-overflow privilege escalation
Google's June 2026 Android Security Bulletin patches CVE-2025-48595, a High-severity integer overflow in the Android Framework component that Google reports is under "limited, targeted exploitation" (Android Security Bulletin, 2026-06-01). The bug gives a local attacker — typically a malicious app already on the device — privilege escalation with no user interaction and no prior privileges, reaching system-level code execution across Android 14, 15, 16 and 16-QPR2 (BleepingComputer, 2026-06-02). The "limited, targeted" descriptor and the Framework location are, in our assessment, consistent with the historical pattern of commercial-spyware operators weaponising Framework LPEs against high-value targets — but no cited source attributes this specific case; the full fix requires reaching the 2026-06-05 patch level, which also carries chipset fixes from Qualcomm, MediaTek, Imagination and Unisoc (Android Security Bulletin, 2026-06-01). Defenders managing Android fleets: push the 2026-06-05 patch level via MDM/EMM and gate non-compliant devices via Security-Patch-Level compliance policy; disable sideloading and restrict installs to managed stores; this is doubly relevant for Swiss federal device fleets given the G7 Évian travel window (§ 1).
CVE Summary Table
A third actively-exploited CVE added to KEV this window — CVE-2022-0492, a Linux cgroup-v1 release_agent container escape — is covered in full in today's deep dive (§ 5).
| CVE | Product | CVSS | EPSS | KEV | Exploited | Patch | Source |
|---|---|---|---|---|---|---|---|
| CVE-2024-21182 | Oracle WebLogic Server, versions 12.2.1.4.0 / 14.1.1.0.0 | 7.5 | high | yes (2026-06-01) | yes — unauth T3/IIOP | Oracle CPU Jul 2024 | THN |
| CVE-2025-48595 | Android Framework (14/15/16/16-QPR2) | High | n/a | yes (2026-06-02) | yes — limited, targeted | 2026-06-05 patch level | Android Bulletin |
| CVE-2022-0492 | Linux kernel cgroup v1 (< 5.17) | 7.0 | n/a | yes (2026-06-02) | yes — container escape | kernel 5.17+ / distro backport | CISA |
3. Research & Investigative Reporting
Sophos finds an attacker-built, AI-orchestrated EDR-evasion testing lab during incident response
Sophos X-Ops disclosed an EDR-evasion development-and-testing environment recovered during an incident-response engagement and linked to an active (unnamed, still-under-investigation) ransomware group (Sophos X-Ops, 2026-06-02). The framework's Python payload generator — many modules partly AI-generated, with Russian-language comments — carried nearly 80 modules covering more than 70 evasion techniques. What distinguishes the lab is its agentic structure: a coordinator agent set rules for role-separated agents (EDR testing, OPSEC hardening, documentation, proxy stress-testing, VM deployment) connected over the Model Context Protocol to a Git repository, with the operator using the Cursor AI IDE and Ludus for rapid VM provisioning (Help Net Security, 2026-06-02). Payloads were tested against three isolated Windows Server 2022 VMs — one Sophos-equipped, one CrowdStrike-equipped, one EDR-free as baseline — with a Sliver/Cobalt Strike C2 stack and a Cloudflare Worker fronting the backend.
Why it matters to us: This is a concrete data point on adversaries operationalising agentic AI for detection-engineering against the exact EDR products (Sophos, CrowdStrike) deployed across CH/EU public-sector estates. The defensive principle is unchanged — the productivity multiplier is on the attacker's tooling, not a new bypass class — but it raises the priority of behavioural telemetry on payload-origin paths: Sophos noted the customer detection fired on "malicious payloads originating from a testing directory," a useful hunt pivot for anomalous build/test artefacts on endpoints.
ANNUAL REPORT — Sophos 2026 Active Adversary Report: identity is the dominant intrusion root cause [SINGLE-SOURCE]
Sophos published its 2026 Active Adversary Report (drawing on 661 IR/MDR cases) on 2026-06-02 (Sophos X-Ops, 2026-06-02). Per PD-9 this report gets one treatment; the findings that change defender priorities rather than the survey scorecard: identity-based compromise — stolen/valid credentials, brute force, and phishing — was the leading root cause, and missing or misconfigured MFA was present in a majority of incidents. Time from initial access to Active Directory compromise has compressed materially, with Impacket among the most frequently observed post-exploitation toolkits and AnyDesk the most-abused legitimate remote-access tool. The recurring telemetry blind spots are the actionable part: firewall logs were missing in roughly half of ransomware cases, and a meaningful share of compromised Windows Servers were running end-of-life builds. [SINGLE-SOURCE] (vendor IR telemetry report).
Why it matters to us: The hunt targets generalise directly to public-sector AD estates — alert on Impacket artefacts (impacket-* tool names in process trees, secretsdump-style NTDS access, SMBExec/WMIExec parent processes), instrument the initial-access-to-DC-compromise window, inventory EOL Windows Servers, and verify firewall log retention before an incident rather than during one.
SANS ISC: SVG phishing wave abuses a non-standard MIME type to slip past WAF/email pattern-matching [SINGLE-SOURCE]
SANS ISC handler Xavier Mertens documented a fresh wave of phishing emails carrying SVG attachments whose embedded JavaScript is obfuscated with combined Base64 + XOR encoding and, on decode, redirects the victim via window.location.href to a credential-harvesting page (SANS ISC, 2026-06-02). The notable evasion is the use of <script type="application/ecmascript"> instead of the standard text/javascript — browsers execute both identically, but email-security and WAF products that pattern-match specifically on text/javascript can miss the non-standard declaration. Because SVGs open natively in Windows browsers, the redirect fires on file open with no extra click. [SINGLE-SOURCE] (SANS Internet Storm Center). Detection: flag email attachments of Content-Type: image/svg+xml that contain embedded <script> elements; treat the application/ecmascript/application/javascript MIME variants as equivalent to text/javascript in inspection rules; sandbox SVG attachments before delivery and watch newly-registered low-cost TLDs (the campaign used a .cfd domain) at the proxy.
Operation XENOFISCAL: SideCopy (APT36) hits provincial treasury officials with XenoRAT via an mshta/HTA chain
Seqrite Labs documented Operation XENOFISCAL, a SideCopy (Transparent Tribe / APT36, Pakistan-attributed) campaign against finance officials across Afghanistan's 34 provincial treasury directorates (Mustoufiats) (Seqrite Labs, 2026-05-29). The chain is the group's long-standing signature — a spear-phishing ZIP carrying a Pashto-language LNK that invokes mshta.exe to pull an obfuscated HTA/JavaScript stage from a compromised education domain, which stages .NET loaders in memory before dropping the publicly available XenoRAT (keylogging, screen capture, remote shell) (The Hacker News, 2026-06-02). Persistence uses a Registry Run key typosquatting Microsoft Edge ("Edgre") plus a Scheduled Task; C2 ran on an EU-hosted bulletproof AS (AS59711) previously tied to the group. ATT&CK: T1566.001, T1218.005 (mshta proxy execution), T1547.001, T1053.005.
Why it matters to us: The victimology is South-Central Asian, but the LNK→mshta.exe→HTA→RAT pattern and the typosquatted-product Run-key persistence are directly transferable hunt content for any public-sector treasury/finance environment: alert on mshta.exe spawning wscript.exe or making outbound HTTP, and on Run-key values that misspell legitimate Microsoft product names.
4. Updates to Prior Coverage
UPDATE: Gamaredon weaponises WinRAR CVE-2025-8088 and adds the GammaSteel stealer
UPDATE (originally covered 2026-06-02): Sekoia TDR's "FSB's Matryoshka" series adds material technical detail to the Gamaredon (UAC-0010 / ACTINIUM) tooling consolidation covered yesterday: the group is exploiting the WinRAR path-traversal flaw CVE-2025-8088 as an initial-access vector, using the traversal to write payloads directly into
%APPDATA%\…\Start Menu\Programs\Startup\for persistence without a Registry or Scheduled-Task artefact (Sekoia TDR, 2026-06-01).The series also names GammaSteel, a modular file-stealer (consolidating prior QuietSieve/HarvesterX-class modules) that captures files by extension and — newly — exfiltrates to attacker-controlled S3-compatible cloud storage in addition to Gamaredon's previously documented HTTP/Telegram channels (The Hacker News, 2026-06-02). The full chain runs WinRAR archive → GammaPhish (HTA) → GammaLoad (VBScript downloader) → GammaWorm/GammaSteel.
Delta for defenders: CVE-2025-8088 is fixed in WinRAR 7.13 (August 2025), so the entry vector is closed by patching — inventory WinRAR versions across the estate. Hunt for archive utilities writing executables or
.vbsintoPrograms\Startuppaths (Sysmon EID 11 on target path containingPrograms\Startup), WinRAR spawningwscript.exe/mshta.exe, and VBScript processes making outbound requests to S3 endpoints inconsistent with normal business traffic. The targeting is Ukraine-centric, but the WinRAR vector reaches any organisation that opens archive-format lures.
Changes since first coverage(1 prior appearance)
- 2026-06-022026-06-02
5. Deep Dive — Linux cgroups v1 release_agent container escape (CVE-2022-0492) re-enters active exploitation
Background. CVE-2022-0492 was disclosed and patched in early 2022; Palo Alto Unit 42 published the canonical technical analysis in March 2022, walking through how the cgroup-v1 release_agent mechanism becomes a container-escape primitive and how earlier mainline kernels shipped without the missing capability check (Unit 42, 2022-03-07). It has sat quietly for four years. CISA's addition of the CVE to the Known Exploited Vulnerabilities catalog on 2026-06-02 (CISA, 2026-06-02) signals fresh in-the-wild exploitation — consistent with attackers harvesting the large tail of unpatched legacy kernels still running container-dense workloads.
The bug. The kernel's cgroup_release_agent_write() handler in kernel/cgroup/cgroup-v1.c failed to verify that a process writing the cgroup-v1 release_agent file holds CAP_SYS_ADMIN in the initial user namespace — CWE-862 Missing Authorization, CVSS 7.0 (Red Hat, CVE-2022-0492). The release_agent is a host path the kernel executes as root on the host whenever the last task leaves a cgroup that has notify_on_release set. Because the write was under-authorised, a process that can mount or reach a writable cgroup-v1 hierarchy can point release_agent at an attacker-controlled script and then empty a cgroup to trigger it — code execution crosses the container boundary into the host root context. This is the textbook T1611 Escape to Host (MITRE ATT&CK T1611).
Exploitation prerequisites — where it actually bites. The attacker needs prior code execution inside a container (initial access via some other vector — an exposed app, a malicious image, a prior § 1-class foothold). The escape requires the ability to mount a cgroup-v1 hierarchy and write release_agent, which in turn requires CAP_SYS_ADMIN in the user namespace that owns that hierarchy. Unit 42 details the dangerous unprivileged path: a process creates a new user namespace (unshare) to obtain CAP_SYS_ADMIN within that namespace and mount a cgroup-v1 hierarchy — viable when the container runtime has not confined the workload with a seccomp profile blocking the mount/unshare calls or an AppArmor/SELinux policy (Unit 42's final write-up scopes the technique to this unprivileged path). A workload directly granted CAP_SYS_ADMIN reaches the same mount/release_agent primitive trivially, which is why over-broad capability grants are dangerous in their own right. The common denominator is a permissively-profiled container: no seccomp, no mandatory-access-control LSM. Self-hosted Kubernetes clusters and bespoke runtimes that strip the default Docker seccomp profile are the typical exposed surface; managed clusters with hardened pod-security defaults and cgroup-v2-only hosts are largely out of scope.
Kill chain. Initial access into the container (T1190/T1610-class) → discovery of cgroup-v1 writability and capability set (T1082) → T1611 Escape to Host: mount a cgroup-v1 controller, set notify_on_release=1, write a host-path payload into release_agent, then spawn-and-exit a process inside a child cgroup so the kernel executes the payload as host root → host-level execution, after which the operator has the usual post-escape options (credential theft from the host, lateral movement to the orchestrator control plane, T1610 deploying further containers).
Hunt and detection concepts (no rule code): the highest-signal artefact is a write to a release_agent file anywhere under /sys/fs/cgroup/** by a non-root or containerised process — Falco/sysdig ship a community rule for exactly this, and auditd can watch the path. Pair it with auditing of in-container mount() of cgroup filesystems and unshare/clone calls that create new user namespaces (auditd syscall rules), and with Linux Sysmon (EID 1) on processes whose executable path was just written via a cgroup release_agent. On the host, a root-context process spawned with no normal parent lineage (kernel-invoked) executing a script from a container-writable path is the escape firing.
Hardening / mitigation. Patch the kernel to 5.17+ or apply the distro backport (every maintained enterprise distro shipped one in 2022) — this restores the CAP_SYS_ADMIN check and closes the class. Independently of patching, the misconfiguration controls neutralise the path: enforce a seccomp profile (the Docker/containerd defaults already block the required mount), apply AppArmor or SELinux confinement to every workload, never grant CAP_SYS_ADMIN to application containers, and move hosts to cgroup v2 exclusively (systemd.unified_cgroup_hierarchy=1), which does not expose the release_agent escape primitive at all. Mounting /sys/fs/cgroup read-only inside containers removes the write target. For Swiss/EU public-sector teams running self-managed Kubernetes or container hosts on long-lived LTS kernels, this KEV addition is the prompt to verify both the kernel patch level and the pod-security/seccomp baseline, since either control alone defeats the escape.
6. Action Items
- Close internet exposure of Oracle WebLogic T3/IIOP and confirm the July 2024 CPU is applied (§ 2, CVE-2024-21182). It is actively exploited unauthenticated; block T3/IIOP at the perimeter, restrict to internal admin subnets via connection filters, and alert on external initiators to ports 7001/7002.
- Push the Android 2026-06-05 patch level across MDM/EMM fleets and gate non-compliant devices for CVE-2025-48595 (§ 2) — prioritise Swiss federal/cantonal devices given the G7 Évian travel window.
- Verify container hosts on two axes for CVE-2022-0492 (§ 5): kernel ≥ 5.17 (or distro backport) and a seccomp/AppArmor/SELinux baseline on every workload; migrate hosts to cgroup-v2-only. Deploy the Falco/
auditdwatch onrelease_agentwrites under/sys/fs/cgroup. - Run G7 Évian readiness for Geneva–Vaud-corridor and Swiss public-sector orgs (§ 1): pre-stage DDoS mitigation, review customer-facing-IdP MFA, rotate admin credentials before 15 June, and brief travelling staff on mobile-device physical security.
- Inventory WinRAR to ≥ 7.13 and hunt Startup-folder writes to close the Gamaredon CVE-2025-8088 entry vector (§ 4); alert on archive utilities writing
.exe/.vbsintoPrograms\Startup. - Move credential-manager and high-value account authentication off TOTP to FIDO2/passkeys (§ 1, Dashlane), and add detection for rapid sequential auth attempts carrying different OTP values from one source.
- Operationalise the Sophos AAR hunt targets (§ 3): alert on
Impacketartefacts (secretsdump/SMBExec/WMIExec), audit AnyDesk use, verify firewall-log retention, and inventory EOL Windows Servers before an incident forces the question.
7. Verification Notes
- Recency window: 36 h (gap to prior daily 2026-06-02 ≈ 24 h — standard daily class; no coverage-window extension required).
- Items dropped (dedup / already covered):
- Operation Dragon Weave (China-aligned RUSTCLOAK → AZUREVEIL/AdaptixC2, Czech Republic/Taiwan; surfaced by S3 and S4) — already covered in full as the 2026-06-02 deep dive; no material in-window delta, dropped.
- KnowledgeDeliver CVE-2026-5426 (Mandiant ViewState/machineKey unauth RCE) — already covered in the 2026-W22 weekly vulnerability roll-up; primary disclosure 2026-05-25 is out-of-window. Dropped.
- Items dropped (out-of-window, PD-7):
- South Staffordshire Water ICO £963,900 Cl0p fine (ZeroLogon CVE-2020-1472, 20-month undetected persistence) — strong water-sector/NIS2 content, but the enforcement action dates to 2026-05-12 (~3 weeks old) with no fresh in-window development; only the enforcement-register listing carried an in-window timestamp. Held out; may resurface if a fresh angle appears.
- Trend Micro Apex One CVE-2026-34926 (directory traversal, agent-update-channel abuse) — vendor/CERT-FR advisories date 2026-05-21/22 (out-of-window) and the CVE is already in
cves_seen; the only in-window hook was the US-FCEB KEV remediation deadline, which is not a fresh-threat signal for this audience (PD-13). Dropped. - Windows DNS Client CVE-2026-41096 (CVSS 9.8) and Hyper-V CVE-2026-40402 (CVSS 9.3) — May 2026 Patch Tuesday flaws (patched 2026-05-12, out-of-window); MSRC assesses exploitation "Unlikely"/"Less Likely" (PoC-only for the DNS flaw), so neither clears a § 2 active-exploitation gate. The in-window NCSC-NL advisory update concerned active exploitation of the companion Netlogon flaw CVE-2026-41089, which was already covered on 2026-06-02. Dropped.
- codexui-android npm OpenAI Codex token theft (Aikido Security) — primary disclosure 2026-05-27 and corroboration 2026-06-01 both fall outside the 36 h window; logged for possible later pickup if fresh reporting appears.
- Items dropped (verification / fake-news guard):
- FSB claim of Western-intelligence spyware on Russian officials' phones — single self-attributing FSB statement (carried by The Record and Meduza, 2026-06-02) with no CVE, sample, or independent technical corroboration; excluded as awareness-only per the fake-news guard. Will be reassessed if technical evidence surfaces.
- Items dropped (periodic-report dedup, PD-9):
- ENISA NIS360 2026 (rail/drinking-water/wastewater enter the risk zone) — the periodic report was already given its dedicated treatment in the 2026-W22 weekly policy section; not re-summarised here.
- Reduced confidence (aggregator-only sourcing): the Dashlane item (§ 1) is backed by TechCrunch, The Hacker News and BleepingComputer — Dashlane's own support advisory returned HTTP 403 and no vendor/regulator primary was reachable. TechCrunch (Zack Whittaker) is original reporting rather than a pure restatement, but all three are journalism hosts; treat the technical specifics as press-sourced pending a vendor advisory.
- Single-source items (flagged in-line): Sophos 2026 Active Adversary Report (§ 3 — single vendor IR-telemetry report; vanity percentages deliberately omitted per PD-4, only structural/hunt findings carried); SANS ISC SVG-phishing diary (§ 3 — sole source SANS Internet Storm Center, HIGH-reliability primary technique research). The § 5 deep dive (CVE-2022-0492) leans on Unit 42 for mechanics plus CISA as the in-window KEV-addition disclosing party (national-CERT carve-out for the exploitation-status fact).
- Contradictions / notes: sub-agents disagreed on the WinRAR CVE-2025-8088 fixed version (7.10 vs 7.13); the brief states 7.13 (August 2025), the version consistent with the vendor's published fix. The unverified "Cobalt Strike / Sodinokibi honeypot payload" detail one sub-agent attached to the WebLogic item was dropped — the cited reporting confirms active exploitation but not those specific payloads.
- Sub-agents: S1–S4 all returned (Claude Sonnet 4.6). No stalls.
- Coverage gaps: databreaches-net (Wayback fallback unusable; persistently unavailable); inside-it-ch (Cloudflare challenge / 403, Wayback unusable); sophos-xops (blog feed HTTP 503 — content recovered via WebSearch + direct article fetch, not a true gap); cert-fr-actu-recent (actualité feed stalled at October 2025); cnil-fr (no in-window enforcement action identified); sec-disclosures-edgar (no qualifying 8-K Item 1.05 cyber-incident filings in window); cisco-psirt, jpcert, apple-security — no in-window items found.