Comparing 22 September 2026 with the previous day 21 September 2026.
This is more than a patch notice: CISA confirms active in-the-wild exploitation of the Zyxel flaw, with attackers gaining command and SYSTEM access, and the item is raised alongside a Veeam vulnerability.
Microsoft initially classified the flaw merely as a spoofing bug with a CVSS score of 6.5, although it actually allows authenticated remote code execution , the real risk is therefore higher than the original score suggests.
Beyond the bare patch information, the advisory shows that mere network access is sufficient to drive the UniFi gateway into a denial-of-service state via an out-of-bounds write.
The report contains no details beyond the pure patch information; there are no indications of active exploitation, affected victim groups, or specific attack techniques.
The report provides no information beyond the plain CVE description, particularly no indication of active exploitation or attack campaigns.
The report describes a pure DoS vulnerability in UniFi gateways with no indication of active exploitation or strategic implications.
The vulnerability requires existing network access and only results in a denial of service on the gateway device; it is therefore an availability risk rather than a compromise risk for exposed UniFi edge devices.
The vulnerability requires network access and only affects certain UniFi gateway devices; exploitation results only in a denial of service, but a gateway failure can disrupt all downstream network connectivity.
The report shows that multiple Chinese APT groups use the same exploit kit for attacks on Chrome and Microsoft, indicating coordinated or shared infrastructure.
The report describes an active North Korean infiltration tactic using fake IT applicants that goes beyond patch management and affects personnel and supply chain integrity.
The report shows that a state-directed Chinese actor is deliberately compromising network and edge devices to exfiltrate data and to position itself for follow-on attacks , considerably more than a plain patch advisory.
Beyond the pure vulnerability description, the report provides no information on active exploitation, exploit availability, or threat actors.
The report contains no information about active exploitation, attackers, or strategic context, and thus does not go beyond the pure vulnerability information.
The BSI warns of critical vulnerabilities in various SAP applications without naming specific CVE IDs, indicating an aggregated security advisory.
This pure CVE report provides no information beyond patch details about active exploitation, threat actors, or affected sectors.
The EvilTokens service shows that AI-driven phishing-as-a-service industrialises authentication-token theft and business email compromise, undermining plain password- and MFA-based protections.
The report points to a critical remotely exploitable code execution vulnerability in OpenSSH for which no CVE details or exploit information are yet available.
The report shows that device-code phishing was operated as an industrialized, AI-assisted service that compromises Microsoft identities via token abuse , an active campaign signal beyond patch information.
Abuse of the legitimate device code flow enables token theft despite MFA, so stolen OAuth tokens remain valid even after a password change.
The report shows that device code phishing against Microsoft 365 was commercialized as a service and remains a viable attack vector even after the seizure of 50 websites and more than 150 domains.
Multiple vulnerabilities in ESXi and vCenter Server allow code execution, security bypass and denial of service at the hypervisor and management layer.
A zero-day exploit that bypasses Windows Defender demonstrates a new evasion technique that is difficult to detect without a patch.
A zero-day in Windows Defender prevents antivirus updates, thereby undermining the protection mechanism itself, which goes beyond a mere patch note.
An unpatched zero-day PoC blocks Defender updates by filling the disk, and the same researcher's earlier tools were already exploited in live attacks, significantly raising the risk of stale signature content.
BSI update on multiple vulnerabilities in Veeam Backup & Replication that allow privilege escalation and file manipulation; specific CVEs or exploitation status are not provided.
The BSI update contains no CVE reference and no exploitability details, which complicates prioritization for affected Linux systems.
The advisory summarizes multiple Linux kernel vulnerabilities without specific CVE IDs or exploitation evidence; it therefore provides no immediately actionable details.
The attack shows that registering external MFA providers in Entra ID itself becomes an attack surface: a privileged account can use it to capture plaintext passwords during regular logins, without any CVE or software vulnerability being required.
The updated BSI advisory bundles multiple kernel vulnerabilities without CVE IDs, making targeted prioritization and patch mapping more difficult.
The BSI report summarises Adobe's bundled patches for Acrobat and Reader; it names no specific CVE IDs and thus provides no added value beyond the patch information itself.
The report names no CVE, no patch status and no active exploitation; code execution requires visiting an attacker-controlled website.
No changes in this category.
No changes in this category.
The report provides a public exploit for an incomplete Windows patch and describes a novel technique for abusing dangling COM object registrations.
The report shows that North Korean APT groups deliberately abuse IT service providers as a stepping stone into their customers' networks, meaning the real risk lies less with the compromised provider than with its clients.
The campaign shows that North Korean actors use fake recruitment and coding-test processes to target IT and engineering professionals and trick them into executing malicious code , a social-engineering vector that can also reach non-crypto engineering teams in DACH organizations.
The flaws allow not only reading but also writing to the NAS, enabling manipulation or persistence on storage systems , patches are available.
The 'PAYLOAD' ransomware uses tampered Group Policy Objects as a distribution mechanism, so a single compromised AD account can trigger infection across an entire domain network.
The BSI WID does not provide CVE IDs or severity ratings; the report is an aggregated patch update for Chrome and Edge.
Microsoft announces the end of SMS first-factor authentication in Entra ID starting February 2027, requiring early migration to passkeys to avoid sign-in disruptions.
The report contains no CVE ID, no CVSS score, and no exploitation status, which complicates prioritization for patch processes.