Kythos AI vulnerability intelligence
Zero-Day Exploits / Vulnerabilities
A server-rendered, source-attributed view of confirmed CISA KEV entries and all NVD CVEs published in the rolling 48-hour UTC window.
Important distinction: CISA KEV means exploitation has been observed; it does not mean every entry was a zero-day. An NVD publication in the past 48 hours does not mean exploitation has occurred. Source records describe third-party products and do not identify Kythos AI customers, contracts, affiliations, or work history.
Confirmed exploitation intelligence
Zero-Day Exploits / Vulnerabilities
Complete CISA Known Exploited Vulnerabilities catalog, ordered by the date CISA added each entry. Every result is known to be exploited; entries are not necessarily zero-days.
Entries 1,177–1,200 of 1,769
CVE-2018-8405: Microsoft DirectX Graphics Kernel Privilege Escalation Vulnerability
Microsoft · DirectX Graphics Kernel (DXGKRNL)
An elevation of privilege vulnerability exists when the DirectX Graphics Kernel (DXGKRNL) driver improperly handles objects in memory.
Required action: Apply updates per vendor instructions.
CVE-2017-0213: Microsoft Windows Privilege Escalation Vulnerability
Microsoft · Windows
Microsoft Windows COM Aggregate Marshaler allows for privilege escalation when an attacker runs a specially crafted application.
Required action: Apply updates per vendor instructions.
CVE-2017-0059: Microsoft Internet Explorer Information Disclosure Vulnerability
Microsoft · Internet Explorer
Microsoft Internet Explorer allow remote attackers to obtain sensitive information from process memory via a crafted web site.
Required action: Apply updates per vendor instructions.
CVE-2017-0037: Microsoft Edge and Internet Explorer Type Confusion Vulnerability
Microsoft · Edge and Internet Explorer
Microsoft Edge and Internet Explorer have a type confusion vulnerability in mshtml.dll, which allows remote code execution.
Required action: Apply updates per vendor instructions.
CVE-2016-7201: Microsoft Edge Memory Corruption Vulnerability
Microsoft · Edge
The Chakra JavaScript scripting engine in Microsoft Edge allows remote attackers to execute remote code or cause a denial of service (memory corruption) via a crafted web site.
Required action: Apply updates per vendor instructions.
CVE-2016-7200: Microsoft Edge Memory Corruption Vulnerability
Microsoft · Edge
The Chakra JavaScript scripting engine in Microsoft Edge allows remote attackers to execute remote code or cause a denial of service (memory corruption) via a crafted web site.
Required action: Apply updates per vendor instructions.
CVE-2016-0189: Microsoft Internet Explorer Memory Corruption Vulnerability
Microsoft · Internet Explorer
The Microsoft JScript nd VBScript engines, as used in Internet Explorer and other products, allow attackers to execute remote code or cause a denial of service (memory corruption) via a crafted web site.
Required action: Apply updates per vendor instructions.
CVE-2016-0151: Microsoft Windows CSRSS Security Feature Bypass Vulnerability
Microsoft · Client-Server Run-time Subsystem (CSRSS)
The Client-Server Run-time Subsystem (CSRSS) in Microsoft mismanages process tokens, which allows local users to gain privileges via a crafted application.
Required action: Apply updates per vendor instructions.
CVE-2016-0040: Microsoft Windows Kernel Privilege Escalation Vulnerability
Microsoft · Windows
The kernel in Microsoft Windows allows local users to gain privileges via a crafted application.
Required action: Apply updates per vendor instructions.
CVE-2015-2426: Microsoft Windows Adobe Type Manager Library Remote Code Execution Vulnerability
Microsoft · Windows
A remote code execution vulnerability exists in Microsoft Windows when the Windows Adobe Type Manager Library improperly handles specially crafted OpenType fonts.
Required action: Apply updates per vendor instructions.
CVE-2015-2419: Microsoft Internet Explorer Memory Corruption Vulnerability
Microsoft · Internet Explorer
JScript in Microsoft Internet Explorer allows remote attackers to execute remote code or cause a denial of service (memory corruption) via a crafted web site.
Required action: Apply updates per vendor instructions.
CVE-2015-1770: Microsoft Office Uninitialized Memory Use Vulnerability
Microsoft · Office
Microsoft Office allows remote attackers to execute arbitrary code via a crafted Office document.
Required action: Apply updates per vendor instructions.
CVE-2013-3660: Microsoft Win32k Privilege Escalation Vulnerability
Microsoft · Win32k
The EPATHOBJ[network address omitted]pprFlattenRec function in win32k.sys in the kernel-mode drivers in Microsoft does not properly initialize a pointer for the next object in a certain list, which allows local users to gain privileges.
Required action: Apply updates per vendor instructions.
CVE-2013-2729: Adobe Reader and Acrobat Arbitrary Integer Overflow Vulnerability
Adobe · Reader and Acrobat
Integer overflow vulnerability in Adobe Reader and Acrobat allows attackers to execute remote code.
Required action: Apply updates per vendor instructions.
CVE-2013-2551: Microsoft Internet Explorer Use-After-Free Vulnerability
Microsoft · Internet Explorer
Use-after-free vulnerability in Microsoft Internet Explorer allows remote attackers to execute remote code via a crafted web site that triggers access to a deleted object.
Required action: Apply updates per vendor instructions.
CVE-2013-2465: Oracle Java SE Unspecified Vulnerability
Oracle · Java SE
Unspecified vulnerability in the Java Runtime Environment (JRE) component in Oracle Java SE allows remote attackers to affect confidentiality, integrity, and availability via Unknown vectors related to 2D
Required action: Apply updates per vendor instructions.
CVE-2013-1690: Mozilla Firefox and Thunderbird Denial-of-Service Vulnerability
Mozilla · Firefox and Thunderbird
Mozilla Firefox and Thunderbird do not properly handle onreadystatechange events in conjunction with page reloading, which allows remote attackers to cause a denial-of-service (DoS) or possibly execute malicious code via a crafted web site.
Required action: Apply updates per vendor instructions.
CVE-2012-5076: Oracle Java SE Sandbox Bypass Vulnerability
Oracle · Java SE
The default Java security properties configuration did not restrict access to the com.sun.org.glassfish.external and com.sun.org.glassfish.gmbal packages. An untrusted Java application or applet could use these flaws to bypass Java sandbox restrictions.
Required action: Apply updates per vendor instructions.
CVE-2012-2539: Microsoft Word Remote Code Execution Vulnerability
Microsoft · Word
Microsoft Word allows attackers to execute remote code or cause a denial-of-service (DoS) via crafted RTF data.
Required action: Apply updates per vendor instructions.
CVE-2012-2034: Adobe Flash Player Memory Corruption Vulnerability
Adobe · Flash Player
Adobe Flash Player contains a memory corruption vulnerability that allows for remote code execution or denial-of-service (DoS).
Required action: The impacted product is end-of-life and should be disconnected if still in use.
CVE-2012-0518: Oracle Fusion Middleware Unspecified Vulnerability
Oracle · Fusion Middleware
Unspecified vulnerability in the Oracle Application Server Single Sign-On component in Oracle Fusion Middleware allows remote attackers to affect integrity via Unknown vectors
Required action: Apply updates per vendor instructions.
CVE-2011-2005: Microsoft Ancillary Function Driver (afd.sys) Improper Input Validation Vulnerability
Microsoft · Ancillary Function Driver (afd.sys)
afd.sys in the Ancillary Function Driver in Microsoft Windows does not properly validate user-mode input passed to kernel mode, which allows local users to gain privileges via a crafted application.
Required action: Apply updates per vendor instructions.
CVE-2010-4398: Microsoft Windows Kernel Stack-Based Buffer Overflow Vulnerability
Microsoft · Windows
Stack-based buffer overflow in the RtlQueryRegistryValues function in win32k.sys in Microsoft Windows allows local users to gain privileges, and bypass the User Account Control (UAC) feature.
Required action: Apply updates per vendor instructions.
CVE-2022-26318: WatchGuard Firebox and XTM Appliances Arbitrary Code Execution
WatchGuard · Firebox and XTM Appliances
On WatchGuard Firebox and XTM appliances, an unauthenticated user can execute arbitrary code.
Required action: Apply updates per vendor instructions.
Complete fixed-window snapshot
CVEs — Past 48 Hours
All unique CVEs returned by NVD for the displayed UTC publication window. The snapshot is accepted only when its normalized unique-item count exactly equals the API totalResults value.
Entries 73–96 of 1,041
CVE-2026-123568
Sveltejs devalue before 5.6.3 contains a prototype pollution vulnerability that allows attackers to alter object prototypes by supplying untrusted data to uneval. When the generated code is later evaluated, it creates objects with polluted prototypes, producing output data whose shape differs from the original input.
CVE-2026-123567
Sveltejs devalue before 5.6.4 contains a prototype injection vulnerability that allows attackers to inject __proto__ own properties into objects returned by devalue.parse and devalue.unflatten. Attackers can supply crafted serialized input that, when downstream code merges results via Object.assign, alters the target object's prototype.
CVE-2026-123566
Sveltejs devalue before 5.9.3 contains a denial of service vulnerability that allows attackers to cause CPU amplification by supplying sparse arrays to uneval. Attackers who influence sparse values passed to uneval can force synchronous work proportional to the declared array length, blocking the event loop.
CVE-2026-123565
Sveltejs devalue before 5.9.3 contains a resource exhaustion vulnerability where repeated primitive strings cause quadratic expansion in the uneval function. Attackers can supply a small crafted payload that is parsed and then passed to uneval, producing a very large serialized string that exhausts memory or CPU.
CVE-2026-123564
Sveltejs devalue from 1.0.0 before 5.9.3 contains a memory exhaustion flaw in uneval, whose generated code for sparse arrays allocates memory proportional to the array's declared length when evaluated. Attackers who can influence a very large sparse array serialized by an application can exhaust memory in…
CVE-2026-123563
Devalue before 5.9.3 contains a validation bypass vulnerability that allows custom ArrayBuffer revivers to skip typed-array allocation checks in the parse function. Attackers can supply very small crafted input processed by a malformed ArrayBuffer reviver to allocate massive ArrayBuffers, causing memory exhaustion and denial of service.
CVE-2026-123562
devalue (sveltejs/devalue) 5.8.0 before 5.9.3 contains an unhandled promise rejection in stringifyAsync: when serializing multiple promises, a later promise rejecting before an earlier one settles leaves an internal rejection unhandled, even if the caller catches the returned promise. An attacker able to influence the timing or…
CVE-2026-123561
devalue before 5.9.3 allows malformed null-prototype object keys to bypass parse's __proto__ rejection through property-key coercion, an incomplete fix for GHSA-mwv9-gp5h-frr4. Attackers can supply crafted payloads that yield objects with a __proto__ own property, which alone does not cause prototype pollution.
CVE-2026-123560
Zebra before 7.0.0 contains an incomplete cleanup vulnerability that allows unauthenticated peers to stall initial sync by serving a forged checkpoint block body. Attackers can supply a body with forged authorizing data near the final checkpoint, leaving rejected hashes in the sent-hash set so the node…
CVE-2026-123559
Zebra before 7.0.0 contains an inefficient algorithmic complexity vulnerability that allows unauthenticated peers to exhaust verification threads by submitting transactions with many transparent inputs. Attackers can relay invalid transactions with tens of thousands of inputs, forcing quadratic ZIP-244 or ZIP-243 digest recomputation that stalls Rayon-backed block…
CVE-2026-123558
Zebra zebrad before 7.0.0 contains a denial of service vulnerability that allows a malicious peer to stall checkpoint sync by cancelling the verifier rewind after a failed commit. Attackers can serve a forged block with altered authorizing data plus a restart-triggering block, leaving the verifier rejecting…
CVE-2026-123557
ZcashFoundation Zebra zebrad before 7.0.0 contains a reachable assertion race condition that allows unauthenticated remote peers to abort the node by triggering disconnection during a blocked outbound send. Malicious peers can return empty FindBlocks or FindHeaders replies and stop reading their socket, dropping an unresolved MustUseClientResponseSender…
CVE-2026-123556
ZcashFoundation Zebra from 4.5.0 before 7.0.0 contains a protection mechanism failure that bans honest peers by scoring mempool lock-time rejections at the ban threshold. Attackers can broadcast transactions locked until the current height right after each block, causing lagging nodes to ban honest relaying peers for…
CVE-2026-123555
Zebra from 2.3.0 before 7.0.0 contains a denial of service vulnerability that allows unauthenticated peers to evade misbehavior scoring by sending transactions or blocks with invalid transparent scripts. Attackers can flood mempool with script-failing transactions or serve forged v5 block bodies with altered scriptSig bytes to…
CVE-2026-123554
Zebra zebrad 6.4.0 before 7.0.0 performs quadratic work in the mempool coinbase-maturity check because Transaction[network address omitted]oinbase_spend_restriction() rebuilds the transaction's output list once per spent outpoint. Any connected peer can repeatedly send unsigned transactions with many transparent inputs and outputs, which are rejected without a misbehavior…
CVE-2026-123553
OpenAM before 16.1.4 ships built-in identities in the Top Level Realm (demo, anonymous, wsc, wsp, SecurityTokenService and agentAuth) with passwords that are identical on every installation, publicly known, and able to authenticate. Unauthenticated remote attackers can log in as these identities to obtain authenticated sessions, and…
CVE-2026-123552
OpenAM before 16.1.4 contains a server-side request forgery vulnerability in certificate CRL checking that fetches CRLs from the distribution point named in a presented certificate without restricting the destination or verifying the downloaded CRL's signature. Attackers holding a trusted-CA-chained certificate can make OpenAM request internal hosts,…
CVE-2026-123551
OpenAM from 14.0.0 before 16.1.4 contains an information disclosure vulnerability that allows unauthenticated attackers to read live session tokens from the notification WebSocket. Attackers can bypass the agent-only check through an unvalidated internal forward on legacy Liberty ID-FF endpoints, collecting iPlanetDirectoryPro tokens to hijack user and…
CVE-2026-123550
OpenAM before 16.1.4 contains an authentication bypass vulnerability in its SAML v2.0 service provider that accepts unsigned assertions on the ECP (PAOS) and HTTP-Artifact bindings when WantAssertionsSigned is false. Unauthenticated remote attackers can submit forged assertions to obtain sessions as any mapped user, including administrators, and…
CVE-2026-123549
OpenAM before 16.1.4 fails to enforce the Valid Creation Attributes list during self-registration, allowing anonymous attackers to set restricted attributes including group membership. Attackers can register through the legacy self-service REST endpoint or User Self-Service flow to join administrative groups and gain full administrative control of…
CVE-2026-123548
OpenIdentityPlatform OpenAM before 16.1.4 contains an open redirect because WS-Federation logout, SAML v2.0 logout and name-identifier management, SAML v2.0 identity provider finder, and some authentication pages do not check destination URLs against allowed lists. Attackers can craft OpenAM-domain links that send users to arbitrary external sites…
CVE-2026-123547
OpenAM before 16.1.4 contains an open redirect vulnerability that allows attackers to redirect users to external sites because RelayState is unchecked when no relayStateUrlList is configured. Attackers can craft single sign-on links with a malicious RelayState to send users from the trusted OpenAM origin to phishing…
CVE-2026-123546
OpenAM before 16.1.4 contains an open redirect and cross-site scripting vulnerability that allows remote attackers to control post-login and logout destinations via the goto and gotoOnFail parameters. Attackers can craft OpenAM login or logout links that redirect victims to phishing sites or run script in the…
CVE-2026-123545
OpenAM from 14.5.1 before 16.1.4 contains a cross-site request forgery vulnerability in the OAuth 2.0 device authorization flow that allows attackers to approve device codes when consent is skipped for implied-consent clients. Attackers can start a device authorization and lure a logged-in user to a crafted…