CVE-20XX-?
agent?

Just published.

The newest CVEs, straight from CVE.org and NVD.

408,384 CVEs · updated 2026-11-19 03:38 UTC

Reset all
  1. CVE-2026-117634

    Pleasanter contains an incorrect authorization vulnerability in Publishing feature. If this vulnerability is exploited, a remote unauthenticated attacker who can access a site published using the product may obtain, by sending a specially crafted request, information on records of unpublished sites (e.g., titles), as well as the names of users, groups, and organizations registered in the same tenant. Note that the vulnerability only affects configurations where both of the following conditions are met: a site is published using the Publishing feature, and the published site contains link items (items linked to other sites, users, groups, or organizations).

    Auth & access Implem Inc. Pleasanter

    6.9 MEDIUM 0.4% exploit odds No fix listed
  2. CVE-2026-121811

    Pleasanter contains an issue with improper limitation of a pathname to a restricted directory ('Path Traversal') in the API. If this vulnerability is exploited, an arbitrary file on the server where the product is running may be obtained by a remote authenticated attacker.

    Disclosure Server Implem Inc. Pleasanter

    5.3 MEDIUM 0.5% exploit odds No fix listed
  3. CVE-2026-125276

    Pleasanter contains an issue with insertion of sensitive information into sent data in the API. If the affected product receives a specially crafted API request sent by a remote authenticated attacker, the authentication credentials of other users, including administrators, such as API keys, may be obtained. As a result, the attacker may perform arbitrary operations on the affected product with the privileges associated with the obtained credentials.

    Disclosure Implem Inc. Pleasanter

    7.1 HIGH 0.6% exploit odds No fix listed
  4. CVE-2026-107949

    A SQL injection vulnerability affects an administrative API in HCL Commerce versions 8, 9.0.1 and 9.1.0 - 9.1.21.  A remote attacker with an administrative role could use a crafted URL to infer unauthorized data from the database.

    Injection Server HCLSoftware Commerce

    6.5 MEDIUM 0.4% exploit odds No fix listed
  5. CVE-2026-93459

    Use of externally-controlled format string in certain ASUS Router firmware allows a nearby administrator to read or write memory, potentially resulting in a denial-of-service, via crafted format specifiers sent to the logging/summary function. Refer to the ' Security Update for ASUS Router Firmware ' section on the ASUS Security Advisory for more information.

    Other Network gear ASUS Router Firmware

    8.5 HIGH 0.2% exploit odds No fix listed
  6. CVE-2026-131820

    A server-side request forgery vulnerability in GitHub Enterprise Server allowed an authenticated user with repository write access to send requests to internal services through Slack webhook validity checks, potentially leading to remote code execution on the appliance. Exploitation required Secret Protection with secret scanning and validity checks for partner patterns enabled. This vulnerability affected all versions prior to 3.23 and was fixed in versions 3.22.5, 3.21.10, and 3.20.12. This vulnerability was reported via the GitHub Bug Bounty program.

    Disclosure Server GitHub Enterprise Server

    7.7 HIGH 0.8% exploit odds Fix listed
  7. CVE-2026-131647

    A CRLF injection vulnerability was identified in GitHub Enterprise Server that allowed a user with write access to a repository to bypass branch protections and update branches or tags in that repository through specially crafted input when resolving pull request conflicts. This vulnerability affected all versions of GitHub Enterprise Server prior to 3.23 and was fixed in versions 3.19.16, 3.20.12, 3.21.10, and 3.22.5. This vulnerability was reported via the GitHub Bug Bounty program.

    Other Server GitHub Enterprise Server

    5.3 MEDIUM 0.5% exploit odds Fix listed
  8. CVE-2026-19399

    Improper Authentication and Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') in ASUS Router firmware allow an unauthenticated nearby user to perform file operations or read files in an unintended directory via crafted HTTP requests during an active user session. Refer to the ' Security Update for ASUS Router Firmware ' section on the ASUS Security Advisory for more information.

    Disclosure Network gear ASUS Router Firmware

    6.1 MEDIUM 0.3% exploit odds No fix listed
  9. CVE-2026-88711

    Improper neutralization of special elements used in an OS command in certain ASUS Router firmware models allows an remote attacker to to execute arbitrary commands with root privileges by inducing a logged-in administrator to visit a malicious website. Refer to the ' Security Update for ASUS Router Firmware ' section on the ASUS Security Advisory for more information.

    Injection Network gear ASUS Router Firmware

    8.7 HIGH 0.6% exploit odds No fix listed
  10. CVE-2026-88760

    Missing Authentication for Critical Function in certain ASUS Router firmware allows unauthenticated nearby users to obtain router information and resources (e.g., multimedia files) via HTTP requests. Refer to the ' Security Update for ASUS Router Firmware ' section on the ASUS Security Advisory for more information.

    Auth & access Network gear ASUS Router Firmware

    5.3 MEDIUM 0.3% exploit odds No fix listed
  11. CVE-2026-93458

    Insertion of sensitive information into sent data in certain ASUS Router firmware allows a remote attacker to obtain router’s UID or login authorization tokens for the victim's linked account via a crafted web page, which requires a logged-in router administrator to actively complete the third-party authorization flow. Refer to the ' Security Update for ASUS Router Firmware ' section on the ASUS Security Advisory for more information.

    Disclosure Network gear ASUS Router Firmware

    6.1 MEDIUM 0.4% exploit odds No fix listed
  12. CVE-2026-116637

    A site administrator could modify files on the appliance host with root privileges and disclose private repository data due to improper path validation in GitHub Enterprise Server. Exploitation required a crafted pre-receive hook environment. This vulnerability affected supported GitHub Enterprise Server versions prior to 3.23 and was fixed in versions 3.19.16, 3.20.12, 3.21.10, and 3.22.5. This vulnerability was reported via the GitHub Bug Bounty program.

    Other Server GitHub Enterprise Server

    8.6 HIGH 0.5% exploit odds Fix listed
  13. CVE-2026-116643

    An SSH certificate authentication bypass vulnerability in GitHub Enterprise Server allowed an authenticated organization member to read and modify another user's private repositories within the trusted certificate authority's scope. Exploitation required an instance using built-in authentication and a valid certificate issued to the attacker by a trusted SSH certificate authority. This vulnerability affected supported GitHub Enterprise Server releases prior to 3.23 and was fixed in versions 3.19.16, 3.20.12, 3.21.10, and 3.22.5. This vulnerability was reported via the GitHub Bug Bounty program.

    Auth & access Server GitHub Enterprise Server

    8.6 HIGH 0.4% exploit odds Fix listed
  14. CVE-2026-119469

    Incorrect default permissions in the ASUS DisplayWidget Center allow a local, low-privileged user to achieve arbitrary code execution in the security context of another user via a crafted executable placed in that user-writable directory, which the application launches when the other user runs it and uses a particular feature.Refer to the ' Security Update for ASUS Display Widget Center ' section on the ASUS Security Advisory for more information.

    Auth & access Network gear ASUS DisplayWidget Center

    7.0 HIGH 0.1% exploit odds No fix listed
  15. CVE-2026-126574

    Missing authorization vulnerability in ASUS Control Center Datacenter and Control Center Enterprise allows remote authenticated users with a low-privileged Viewer role to change the password of an Administrator account, thereby obtaining administrative privileges, via a crafted password update request.  Refer to the ' Security Update for ASUS Control Center  ' section on the ASUS Security Advisory for more information.

    Auth & access Network gear ASUS Control Center Datacenter

    9.4 CRITICAL 0.4% exploit odds No fix listed
  16. CVE-2026-110307

    An incorrect authorization vulnerability was identified in GitHub Enterprise Server that allowed an authenticated user whose numeric user ID matched an enterprise's numeric ID to manage that enterprise's GitHub Apps and generate private keys without enterprise membership. GitHub App ownership checks compared numeric IDs without checking the owner's account type. Exploitation required control of a session for the colliding user account and an enterprise-owned GitHub App. A generated private key could be used to authenticate as the App and access repositories where it was installed, within its granted permissions. On standard fresh installations, the documented collision involved the non-loginable built-in ghost account; the reported GHES reproduction used site administrator impersonation. This vulnerability affected supported GitHub Enterprise Server releases prior to 3.23 and was fixed in versions 3.19.16, 3.20.12, 3.21.10, and 3.22.5. This vulnerability was reported via the GitHub Bug Bounty program.

    Other Server GitHub Enterprise Server

    7.4 HIGH 0.5% exploit odds Fix listed
  17. CVE-2026-106096

    A flaw was found in CRI-O. When the optional Linux LinkLogs feature is enabled, the host pod log directory is bind-mounted into a pod emptyDir and is intended to be read-only, but Linux ignores MS_RDONLY on the initial bind and the CRI mount stays writable. A non-privileged UID 0 container can place a symlink at a future kubelet log path so that privileged conmon later appends attacker-controlled log data to an existing host file. This is an append-only host integrity issue, not universal remote code execution; LinkLogs is not enabled by default.

    Other Server Red Hat Confidential Compute Attestation

    6.4 MEDIUM 0.2% exploit odds No fix listed
  18. CVE-2026-112115

    An untrusted search path in ASUS Armoury Crate allows a local, low-privileged user to achieve privilege escalation by executing a crafted library at a high integrity level, via user-controlled environment settings inherited by an elevated process. Refer to the ' Security Update for Armoury Crate App ' section on the ASUS Security Advisory for more information.

    Code execution Network gear ASUS Armoury Crate

    8.5 HIGH 0.2% exploit odds No fix listed
  19. CVE-2026-112117

    A Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') in ASUS System Control Interface , allows a local, low-privileged user to gain full control over arbitrary existing files and folders outside the allowed directories via a crafted path containing dot-segment sequences that satisfies a folder prefix check, which can lead to unauthorized read and modification of protected files and may result in local privilege escalation. Refer to the ' Security Update for ASUS System Control Interface  ' section on the ASUS Security Advisory for more information.

    Disclosure Network gear ASUS System Control Interface

    7.3 HIGH 0.1% exploit odds No fix listed
  20. CVE-2026-114202

    Use of Hard-coded Cryptographic Key in ASUS Control Center Express allows a remote user to gain administrative privileges via a forged authentication token. Refer to the ' Security Update for ASUS Control Center Express Agent  ' section on the ASUS Security Advisory for more information.

    Auth & access Network gear ASUS Control Center Express

    10.0 CRITICAL 0.7% exploit odds No fix listed
  21. CVE-2026-114204

    Use of Hard-coded Cryptographic Key in ASUS Control Center Express allows a remote user to bypass authentication by forging a session token for an account with a recent valid login, resulting in administrative control of the management console and of the clients it manages. Refer to the ' Security Update for ASUS Control Center Express Agent  ' section on the ASUS Security Advisory for more information.

    Auth & access Network gear ASUS Control Center Express

    9.0 CRITICAL 0.6% exploit odds No fix listed
  22. CVE-2026-114205

    Incorrect authorization in ASUS Control Center Express allows a remote authenticated user with a low-privileged role to create an administrator account, resulting in administrative control of the management console and of the clients it manages. Refer to the ' Security Update for ASUS Control Center Express Agent  ' section on the ASUS Security Advisory for more information.

    Auth & access Network gear ASUS Control Center Express

    9.4 CRITICAL 0.4% exploit odds No fix listed
  23. CVE-2026-105929

    A path traversal vulnerability (CWE-22) was found in ansible-core's ansible-galaxy role installation. An explicit role "name" taken from a role's meta/requirements.yml is used to construct the on-disk installation path without validating that the resulting destination stays within the selected roles directory. A name containing "../" escapes that directory, and an absolute name replaces it entirely. Because ansible-galaxy processes transitive dependencies from an installed role's meta/requirements.yml, a malicious role publisher can cause attacker-supplied files to be written outside the operator's chosen roles directory when the operator installs the role, and, when --force is used, can cause an existing role directory outside that path to be removed and replaced. The impact is bounded by the invoking user's filesystem permissions and requires the operator to install the attacker-influenced content.

    Disclosure Red Hat Migration Toolkit For Applications 8

    8.1 HIGH 0.5% exploit odds No fix listed
  24. CVE-2026-111801

    An authorization bypass through user-controlled key vulnerability was identified in GitHub Enterprise Server that allowed authenticated users to read review-thread data from private repositories they could not access. An attacker required an authenticated account, a readable repository and pull request to use as the request context, and target review-thread identifiers. The thread-preview endpoint resolved supplied identifiers without restricting them to the authorized pull request, exposing private file paths and, when positioning and diff data were available, source-code excerpts, commit identifiers, and line metadata without victim interaction. This vulnerability affected supported GitHub Enterprise Server releases in the 3.21, 3.22 series and was fixed in versions 3.21.10, 3.22.5. This vulnerability was reported via the GitHub Bug Bounty program.

    Auth & access Server GitHub Enterprise Server

    7.1 HIGH 0.4% exploit odds Fix listed
  25. CVE-2026-112740

    A command injection vulnerability was identified in GitHub Enterprise Server that allowed an authenticated site administrator to execute arbitrary commands as the host git user on peer Git server nodes outside the pre-receive hook sandbox by supplying a crafted pre-receive environment download URL. Exploitation required a clustered deployment using custom pre-receive hook environments and at least one peer Git server node. This vulnerability affected all versions of GitHub Enterprise Server prior to 3.23 and was fixed in versions 3.22.5, 3.21.10, 3.20.12, and 3.19.16. This vulnerability was reported via the GitHub Bug Bounty program.

    Injection Server GitHub Enterprise Server

    7.4 HIGH 2% exploit odds Fix listed
  26. CVE-2026-92182

    Netgate pfSense CE XMLRPC Incorrect Authorization Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of Netgate pfSense. Authentication is required to exploit this vulnerability. The specific flaw exists within the XMLRPC endpoint. The issue results from incorrect authorization prior to allowing access to functionality. An attacker can leverage this vulnerability to execute code within the context of the root. Was ZDI-CAN-32353.

    Auth & access Netgate pfSense

    8.8 HIGH 3% exploit odds Fix listed
  27. CVE-2026-131852

    OwnTracks Recorder before 1.0.5 contains a format string vulnerability in json_response() and related output functions that allows unauthenticated attackers to execute code by storing a malicious tid value. Attackers can publish locations via POST /pub or MQTT with printf specifiers in tid, then read them back through GET /api/0/last to leak addresses and write memory.

    Other OwnTracks Recorder

    9.3 CRITICAL 0.8% exploit odds Fix listed
  28. CVE-2026-131853

    Ettercap through 0.8.5.1 contains a heap out-of-bounds read vulnerability in dissector_ssh() in src/dissectors/ec_ssh.c that copies a peer-supplied data_len without checking remaining packet bytes. A malicious SSHv1 client or server in an intercepted session can send a short packet with a large data_len to disclose up to 64 bytes of adjacent heap memory through captured credentials in the interface and log.

    Memory safety Server Ettercap

    6.0 MEDIUM 0.3% exploit odds Fix listed
  29. CVE-2026-131854

    Ettercap through 0.8.5.1 contains a heap out-of-bounds read vulnerability in dissector_mysql() in src/dissectors/ec_mysql.c that copies seed and password bytes without bounds checks. A malicious MySQL server or client can send truncated greeting or login packets to leak adjacent heap memory through hex-encoded output in the interface and log.

    Memory safety Server Ettercap

    7.1 HIGH 0.3% exploit odds Fix listed
  30. CVE-2026-131855

    Ettercap through 0.8.5.1 contains a heap out-of-bounds read vulnerability in dissector_vnc() in src/dissectors/ec_vnc.c that checks total packet length instead of remaining bytes. A malicious VNC server can send the authentication type and challenge in one short packet to leak up to 3 bytes of adjacent heap memory via the logged challenge.

    Memory safety Server

    5.3 MEDIUM 0.3% exploit odds Fix listed
  31. CVE-2026-131856

    Pika before 1.5.0 contains an uncontrolled resource allocation vulnerability that allows malicious brokers to exhaust client memory by declaring oversized frame payloads. Attackers can send frame headers declaring up to 4 GiB payloads, unchecked against frame_max, or very large message bodies to terminate consuming applications.

    Denial of service Pika

    6.9 MEDIUM 0.6% exploit odds Fix listed
  32. CVE-2026-131857

    Supergateway 4.2.0 through 4.2.2 contains a denial of service vulnerability in stdio-to-SSE mode that allows unauthenticated attackers to crash the gateway via prototype-inherited sessionId values. Attackers can send a POST message request with sessionId set to toString or constructor, triggering an unhandled TypeError that terminates every attached MCP session.

    Denial of service supercorp-ai Supergateway

    8.7 HIGH 0.7% exploit odds Fix listed
  33. CVE-2026-131846

    Poppler 0.76.1 through 26.11.0 contains a heap-based buffer overflow in SplashOutputDev::iccTransform() that allows attackers to write out of bounds when rendering in splashModeDeviceN8. Attackers can supply a PDF with an ICCBased image whose alternate is DeviceN, causing getDeviceNLine() to overflow the per-row buffer, crashing the renderer and corrupting heap memory.

    Memory safety Freedesktop Poppler

    5.8 MEDIUM 0.2% exploit odds No fix listed
  34. CVE-2026-131847

    Poppler through 26.11.0 contains a use-after-free vulnerability in NSSCryptoSignBackend.cc that allows attackers to trigger reads of freed CMS arena memory during repeated signature verification. Attackers can supply a signed PDF embedding the signer certificate to crash applications or pdfsig runs validating multiple signatures, or cause incorrect certificate validation results.

    Memory safety Web app

    4.8 MEDIUM 0.1% exploit odds No fix listed
  35. CVE-2026-131848

    Snap7 through 1.4.3 contains an out-of-bounds write vulnerability in TSnap7MicroClient::opReadMultiVars() that allows malicious PLCs to overflow client buffers via unchecked DataLength fields. Attackers controlling or impersonating a PLC can return crafted ReadVar responses to Cli_ReadMultiVars() callers, writing up to 65535 bytes per item to corrupt memory and crash clients.

    Memory safety Device Davenardella snap7

    8.3 HIGH 0.5% exploit odds No fix listed
  36. CVE-2026-131849

    Snap7 through 1.4.3 contains a heap-based buffer overflow in TSnap7MicroClient::opReadSZL() that allows malicious PLCs to overwrite client memory by sending oversized multi-slice SZL responses. A rogue or impersonated PLC can keep flagging more data with attacker-controlled DLen slices, overflowing the 65536-byte opData buffer via Cli_ReadSZL() and Cli_ReadSZLList(), crashing the client and corrupting heap memory.

    Memory safety Device Davenardella snap7

    8.3 HIGH 0.6% exploit odds Fix listed
  37. CVE-2026-131850

    ILIAS 11.0 before 11.6 contains an sql injection vulnerability in RegistrationCodeRepository::getCodesData() that concatenates unvalidated sort fields into an ORDER BY clause. Authenticated users with read access to the Authentication and Registration administration node can supply SQL expressions as the sort parameter to extract database contents via boolean-based inference.

    Injection Server ILIAS-eLearning e.V. ILIAS

    8.6 HIGH 0.7% exploit odds Fix listed
  38. CVE-2026-131851

    ILIAS before 9.25, 10.13, and 11.6 contains an SQL injection vulnerability in ilPRGAssignmentFilter::toConditions() that allows authenticated users to inject SQL through the StudyProgramme members list Name and Status filters. Attackers holding only the OrgUnit permission to view members can break out of LIKE literals and execute stacked statements, such as overwriting the root password hash to gain administrative control.

    Injection Web app

    8.7 HIGH 0.6% exploit odds Fix listed
  39. CVE-2026-131842

    Poppler pdfdetach 24.03.0 through 26.11.0 contains a path traversal vulnerability in utils/pdfdetach.cc that allows attackers to write files outside the output directory by bypassing a string-prefix containment check. Attackers can craft PDFs with embedded file names containing '..' sequences or absolute paths, so -saveall or -savefile extraction creates or overwrites files with the invoking user's privileges.

    Disclosure Device Freedesktop Poppler

    5.7 MEDIUM 0.2% exploit odds Fix listed
  40. CVE-2026-131843

    Poppler pdftohtml through 26.11.0 contains a cross-site scripting vulnerability in HtmlLink::getLinkStart() in utils/HtmlLinks.cc that writes PDF link annotation URIs into href attributes without escaping in HTML output modes. Attackers can craft a PDF whose /URI entry contains a double quote followed by HTML markup or event handlers to execute script in browsers viewing the generated HTML.

    XSS

    5.3 MEDIUM 0.3% exploit odds No fix listed
  41. CVE-2026-131844

    Poppler through 26.11.0 contains an uncontrolled memory allocation vulnerability in the libcurl-based remote file loader that trusts unvalidated HEAD Content-Length values in CurlCachedFileLoader::init(). Malicious HTTP servers can return an enormous Content-Length so CachedFile resizes its chunk cache accordingly, aborting pdfinfo or PDFDocFactory-based applications with std::bad_alloc for denial of service.

    Denial of service Freedesktop Poppler

    7.1 HIGH 0.5% exploit odds No fix listed
  42. CVE-2026-131845

    Poppler 21.01.0 through 26.11.0 contains a time-of-check time-of-use race condition in FormWidgetSignature::signDocument() that allows local attackers to obtain valid signatures over altered content. Attackers with write access to the output file or directory can overwrite or swap the PDF between writing and hashing, causing the signer's key to sign attacker-controlled content.

    Other Freedesktop Poppler

    4.1 MEDIUM 0.1% exploit odds No fix listed
  43. CVE-2026-130600

    Insertion of Sensitive Information Into Sent Data vulnerability in BdThemes Element Pack Elementor Addons bdthemes-element-pack-lite allows Retrieve Embedded Sensitive Data.This issue affects Element Pack Elementor Addons: from n/a through 10.0.9.

    Disclosure Web app BdThemes Element Pack Elementor Addons

    4.3 MEDIUM 0.3% exploit odds No fix listed
  44. CVE-2026-129069

    Missing Authorization vulnerability in John-Michael L'Allier Create mediavine-create allows Exploiting Incorrectly Configured Access Control Security Levels.This issue affects Create: from n/a through 3.1.3.

    Auth & access John-Michael L'Allier Create

    5.3 MEDIUM 0.3% exploit odds No fix listed
  45. CVE-2026-129891

    IO::Socket::SSL::PublicSuffix versions through 2.099 for Perl let an on-path attacker choose the installed public suffix list because update_self_from_url fetches it over cleartext HTTP with no integrity check. Called with no argument, as the module's documentation shows, the updater fetches the list from the `http://` address in the URL constant and checks only that the request succeeded, with no signature or digest over the response. The lines that come back are written into the module's own source file as the built-in list. An attacker on the network path when the updater runs chooses the suffix boundary the module reports to every caller on the host, and that answer persists in the installed module until it is reinstalled. Dropping one rule makes the hostname check in IO::Socket::SSL accept a wildcard certificate that spans the whole public suffix that rule named.

    Crypto Library

    7.5 HIGH 0.1% exploit odds Fix listed
  46. CVE-2026-129892

    IO::Socket::SSL::PublicSuffix versions through 2.099 for Perl allow code injection via a fetched list line that closes the built-in data heredoc when update_self_from_url rewrites the module's source. The rewrite copies the module's source down to the built-in list's heredoc and appends each fetched line below it. Nothing checks that a line looks like a domain. A line that is not a comment loses only what follows a `//` and its trailing whitespace, so a line reading END_BUILTIN_DATA is appended as it stands and ends the data block in the rewritten file. Every fetched line after it is Perl. Arbitrary Perl runs in every process that loads IO::Socket::SSL, which loads this module unconditionally, until the file is reinstalled. The attacker must control the fetched list, through the cleartext default URL or a caller-supplied one, and the updater must run with write access to the installed module.

    Injection Library

    9.8 CRITICAL 0.2% exploit odds Fix listed
  47. CVE-2026-129040

    Missing Authorization vulnerability in WP User Manager WP User Manager wp-user-manager allows Exploiting Incorrectly Configured Access Control Security Levels.This issue affects WP User Manager: from n/a through 2.10.4.

    Auth & access WP User Manager

    5.3 MEDIUM 0.3% exploit odds No fix listed
  48. CVE-2026-129049

    Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in WP Go Maps WP Go Maps wp-google-maps allows Stored XSS.This issue affects WP Go Maps: from n/a through 10.1.12.

    XSS WP Go Maps

    6.5 MEDIUM 0.2% exploit odds No fix listed
  49. CVE-2026-126975

    llama.cpp GGUF File Parsing Heap-based Buffer Overflow Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of llama.cpp. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file. The specific flaw exists within the parsing of GGUF files. The issue results from the lack of proper validation of the length of user-supplied data prior to copying it to a heap-based buffer. An attacker can leverage this vulnerability to execute arbitrary code in the context of the current process. Was ZDI-CAN-34558.

    Memory safety llama.cpp

    7.8 HIGH 0.3% exploit odds Fix listed
  50. CVE-2026-126976

    llama.cpp MTMD Multimodal GGUF File Parsing Integer Overflow Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of llama.cpp. User interaction is required to exploit this vulnerability in that the target must open a malicious file. The specific flaw exists within the parsing of audio multimodal projector GGUF files. The issue results from the lack of proper validation of user-supplied data, which can result in an integer overflow before allocating a buffer. An attacker can leverage this vulnerability to execute arbitrary code in the context of the current process. Was ZDI-CAN-30980.

    Memory safety llama.cpp

    7.0 HIGH 0.2% exploit odds Fix listed

The newest published CVE records from CVE.org and NVD, refreshed every few minutes, with severity, exploit odds and fix status.