CVE-20XX-?
agent?

CVEs of 2026.

Every CVE published in 2026 that we hold, ranked by exploit odds.

644 CVEs · updated 2026-11-19 02:51 UTC

Reset all
  1. CVE-2026-123854

    In JetBrains PyCharm before 2026.3 code execution was possible in Safe Mode via a crafted Jupyter notebook widget module

    Code execution Library

    5.5 MEDIUM 0.2% exploit odds No fix listed
  2. CVE-2026-113775

    A maliciously crafted dynamic library, when loaded from an arbitrary path through the TextCommands.LoadDll command exposed by the Autodesk Fusion API, can trigger an Exposed Dangerous Function vulnerability in the Autodesk Fusion desktop application. A malicious actor may leverage this vulnerability to execute arbitrary code in the context of the current process.

    Code execution

    7.8 HIGH 0.2% exploit odds No fix listed
  3. CVE-2026-115762

    draw.io is a configurable diagramming and whiteboarding application. Prior to 31.1.5, the GraphML decoder contains a prototype-pollution vulnerability. Attacker-controlled GraphML reference keys can address inherited object properties, including prototype objects, and the decoder can then assign attacker-controlled attributes to those objects. Importing a malicious GraphML file can trigger client-side prototype pollution. In the advisory-supported cross-site scripting chain, the editor is opened in Trello mode and the user invokes the Trello integration after the import, allowing the polluted state to execute script in the draw.io origin; an existing Trello connection is not required. This vulnerability is fixed in 31.1.5.

    Code execution Web app

    6.1 MEDIUM 0.2% exploit odds Fix listed
  4. CVE-2026-101352

    A document can carry its own user interface customizations, such as keyboard shortcuts, menus and toolbars, which take effect while the document is open. A command bound in these customizations could run a macro or a script even when the macro security settings did not allow the macros of the document, so a user who pressed a shortcut or chose a menu item in a crafted document could run code without any warning. In fixed versions the customizations of a document are treated like its other macro content, and the macro and script commands in them are only kept when the macros of the document are allowed.

    Code execution

    8.5 HIGH 0.2% exploit odds No fix listed
  5. CVE-2026-125958

    Download of code without integrity check vulnerability in N-able N-central allows Remote Code Inclusion. This issue affects N-central: through 24.6.1.

    Code execution

    7.5 HIGH 0.2% exploit odds No fix listed
  6. CVE-2026-91093

    A potential security vulnerability has been identified in HP One Agent for certain HP PC products, which might allow for escalation of privilege and/or arbitrary code execution. HP is releasing software updates to mitigate the potential vulnerability.

    Code execution

    8.5 HIGH 0.2% exploit odds No fix listed
  7. CVE-2026-15707

    An uncontrolled search path vulnerability (CWE-427) in statically linked OpenSSL libraries within Netskope Client (stAgentSvc.exe) and Endpoint DLP (epdlp.exe) on Windows allows an authenticated local attacker to execute arbitrary code with elevated SYSTEM privileges. The services retain internal build-time paths referencing a secondary drive location. On Windows systems where default volume permissions allow standard users to create root-level directories on secondary drives, an attacker can place a malicious configuration file (openssl.cnf) that is loaded under the NT AUTHORITY\SYSTEM context upon service startup or system reboot. Under default operating system permissions, macOS and Linux are not susceptible, and mobile platforms are unaffected.

    Code execution OS & kernel

    8.5 HIGH 0.2% exploit odds No fix listed
  8. CVE-2026-116500

    IBM Qiskit SDK 2.4.0 through 2.6.1 could allow a local attacker to cause a denial of service due to unbound recursion during deserialization of untrusted QPY payloads.

    Code execution Library

    6.2 MEDIUM 0.2% exploit odds No fix listed
  9. CVE-2026-59172

    Joker is a small Clojure interpreter, linter and formatter written in Go. Prior to 1.8.2, `joker --lint <file>` walks upward from the linted file to locate a `.jokerd/` directory and executes matching `linter.*` files before linting. Those files contain executable Joker/Clojure code, so linting a file inside an untrusted repository can execute repository-supplied code with the linter user's privileges. Editor integrations and CI jobs that automatically lint checked-out source can trigger this behavior. The patch stops executing project-local linter customizations and loads executable customization only from the user's home directory. This issue is fixed in version 1.8.2.

    Code execution Library

    7.8 HIGH 0.2% exploit odds Fix listed
  10. CVE-2026-49028

    Deserialization of untrusted data for some Intel(R) Neural Compressor software before version 3.8 within Ring 3: User Applications may allow a denial of service. Unprivileged software adversary with an authenticated user combined with a low complexity attack may enable denial of service. This result may potentially occur via local access when attack requirements are not present without special internal knowledge and requires no user interaction. The potential vulnerability may impact the confidentiality (none), integrity (low) and availability (low) of the vulnerable system, resulting in subsequent system confidentiality (none), integrity (none) and availability (none) impacts.

    Code execution

    4.8 MEDIUM 0.2% exploit odds No fix listed
  11. CVE-2026-50177

    Deserialization of untrusted data for some Intel(R) Neural Compressor software before version 3.8 within Ring 3: User Applications may allow a denial of service. Unprivileged software adversary with an authenticated user combined with a low complexity attack may enable denial of service. This result may potentially occur via local access when attack requirements are not present without special internal knowledge and requires no user interaction. The potential vulnerability may impact the confidentiality (none), integrity (low) and availability (low) of the vulnerable system, resulting in subsequent system confidentiality (none), integrity (none) and availability (none) impacts.

    Code execution

    4.8 MEDIUM 0.2% exploit odds No fix listed
  12. CVE-2026-42917

    Deserialization of untrusted data for some Intel(R) Neural Compressor software before version 3.8 within Ring 3: User Applications may allow a denial of service. Unprivileged software adversary with an authenticated user combined with a low complexity attack may enable denial of service. This result may potentially occur via local access when attack requirements are not present without special internal knowledge and requires no user interaction. The potential vulnerability may impact the confidentiality (none), integrity (low) and availability (low) of the vulnerable system, resulting in subsequent system confidentiality (none), integrity (none) and availability (none) impacts.

    Code execution

    4.8 MEDIUM 0.2% exploit odds No fix listed
  13. CVE-2026-118627

    MLflow SHAP load_explainer Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of MLflow. 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 loading of SHAP explainer models. The issue results from the lack of proper validation of user-supplied data, which can result in deserialization of untrusted data. An attacker can leverage this vulnerability to execute code in the context of the current user. Was ZDI-CAN-30301.

    Code execution

    7.8 HIGH 0.2% exploit odds No fix listed
  14. CVE-2026-107090

    An issue in Maltego Technologies GmbH Maltego Windows Installer 4.13.0 allows a local attacker to execute arbitrary code via the Maltego Windows installer executable-resolution logic

    Code execution

    7.8 HIGH 0.2% exploit odds No fix listed
  15. CVE-2026-104349

    Creative Cloud Desktop is affected by an Uncontrolled Search Path Element vulnerability that could result in arbitrary code execution in the context of the current user. A low-privileged attacker could exploit this vulnerability to execute arbitrary code. Exploitation of this issue requires user interaction in that a victim must open a malicious file.

    Code execution

    7.3 HIGH 0.2% exploit odds No fix listed
  16. CVE-2026-114353

    Hidden functionality (undocumented backdoor) in the root-privileged 'assistant' daemon in DJI CrystalSky ZS600C firmware v0.3.9.7 through v0.6.6.9 on Android allows a physically-present attacker with USB access to execute arbitrary OS commands as root by sending the hard-coded magic string 'dji26656677' to the USB-gadget serial interface (/dev/ttyGS0), after which each subsequent line is passed to system().

    Code execution Device

    7.0 HIGH 0.2% exploit odds No fix listed
  17. CVE-2026-114096

    Dulwich from 0.24.0 before 1.2.8 contains a symlink following vulnerability in porcelain.checkout(paths=[...]) that allows attackers to write arbitrary files by abusing raw os.open without O_NOFOLLOW. Attackers can craft a repository where a worktree symlink points to .git/hooks/post-checkout, so restoring the file from a tag writes a malicious hook for code execution.

    Code execution Library

    7.3 HIGH 0.2% exploit odds Fix listed
  18. CVE-2026-114197

    Deserialization of untrusted data in Mandiant FLARE capa releases before 10.0.0 allows arbitrary code execution when a crafted .viv or .exe.viv workspace is passed to the capa utility

    Code execution

    8.7 HIGH 0.2% exploit odds No fix listed
  19. CVE-2026-114198

    Deserialization of untrusted data in Mandiant FLARE floss releases before 3.2.0 allows arbitrary code execution when a crafted .viv or .exe.viv workspace is passed to the floss utility

    Code execution

    8.7 HIGH 0.2% exploit odds No fix listed
  20. CVE-2026-101327

    A maliciously crafted DLL placed in a predictable, attacker-writable temporary directory used by the Autodesk Installer can lead to privilege escalation to NT AUTHORITY\SYSTEM due to insufficient validation of loaded libraries. An attacker with local and low-privilege access could exploit this vulnerability to execute arbitrary code as SYSTEM.

    Code execution

    7.8 HIGH 0.2% exploit odds No fix listed
  21. CVE-2026-21175

    Incorrect permission assignment for critical resource in Samsung Update prior to version 3.0.111.0 allows local attackers to execute arbitrary code.

    Code execution Device

    8.6 HIGH 0.2% exploit odds No fix listed
  22. CVE-2026-62847

    Metalama is a Roslyn-based open-source patterns and architecture toolkit for C#. Prior to 2025.1.18, 2026.0.25 and 2026.1.19, Metalama.Framework and Metalama.Tool load and execute compile-time assemblies, auxiliary executable tools and embedded helper assemblies from predictable temporary directories. On `Linux` and `macOS`, the default `Path.GetTempPath()` location is world-writable, and existing artifacts are trusted based on sentinel or `.completed` marker files rather than integrity verification. A local unprivileged user on a shared machine or CI/build agent can pre-create the deterministic directory structure and plant a malicious DLL with the matching sentinel. When a higher-privileged developer or build account subsequently builds, Metalama skips regeneration and executes the planted artifact in the compiler, IDE or MSBuild process as the building user. This permits cross-user privilege escalation, reading source and secrets, supply-chain tampering with build output, or disrupting builds. Default Windows per-user temporary directories are protected by ACLs and are not affected. Deployer-selected insecure cache locations are outside the vulnerability's scope. This issue is fixed in versions 2025.1.18, 2026.0.25 and 2026.1.19.

    Code execution Library

    8.2 HIGH 0.2% exploit odds Fix listed
  23. CVE-2026-21152

    Improper input validation in libMyFilterPlugin.camera.samsung.so prior to SMR Nov-2026 Release 1 allows local attackers to potentially execute arbitrary code. User interaction is required for triggering this vulnerability.

    Code execution Device

    7.0 HIGH 0.1% exploit odds No fix listed
  24. CVE-2026-95614

    The host configuration watcher in Zenitel Connect before 1.11.0 applies a rsyslog configuration file written to a data volume that is shared with the hiatus application container, which allows local attackers with write access to the hiatus container filesystem to execute arbitrary OS commands as root on the host operating system via a crafted rsyslog configuration file.

    Code execution Server

    9.3 CRITICAL 0.1% exploit odds No fix listed
  25. CVE-2026-88982

    Kubeflow Pipelines enables users to build and deploy portable, scalable machine learning workflows. Prior to 2.18.0, the legacy V1/raw-Argo execution cache mutating webhook and cache-server storage path hash selected template fields without the Kubernetes namespace, use the key alone for lookups and duplicate checks, and trust the mutable pipelines.kubeflow.org/execution_cache_key annotation when publishing outputs. In multi-user deployments where tenant namespaces share the cache service, a tenant with workload-submission privileges can read another namespace's cached outputs or seed matching outputs that are substituted into a later step. Standard KFP V2 component pods bypass this cache, matching requires the selected template fields and staleness policy, and arbitrary code execution or service availability impact is not established. This issue is fixed in version 2.18.0.

    Code execution Server

    5.4 MEDIUM 0.1% exploit odds Fix listed
  26. CVE-2026-126697

    The Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4 was distributed with a hidden update mechanism that fetched Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4 updates from its author's own server instead of WordPress.org, which was used to push a backdoored build to sites, allowing the attacker controlling that server to inject arbitrary content into pages and to deliver arbitrary code to affected sites. Version 5.2.1, as published on WordPress.org, bundled the Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4 Update Checker library (updater/ directory) configured to poll https://anadnet.com/updates/?action=get_metadata&slug=quick-pagepost-redirect-Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4. In March 2021 that server delivered a modified 5.2.3 build, different from the 5.2.3 later published on WordPress.org, which kept the update channel and added a the_content filter that, for logged-out visitors on single posts and pages, fetched content from https://w.anadnet.com/bro/3/ (passing the site host, request URI and visitor User-Agent) and prepended the response to the post (observed serving SEO spam links to search engine crawlers). Because the version string matched a later WordPress.org release, affected sites were not offered a clean update until 5.2.4. Indicators of compromise: - Outbound requests to anadnet.com or w.anadnet.com - An updater/ directory (Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4 Update Checker, Puc v4p10) inside wp-content/Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4/quick-pagepost-redirect-Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4/ - The strings filter_the_content_in_the_main_loop, ppr_pro, QQQEND or w.anadnet.com in page_post_redirect_Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4.php - page_post_redirect_Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4.php with MD5 ad717da18cf8a2b69899c0d7dafee05a (tampered 5.2.3) Remediation: remove the Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4 (it has been closed on WordPress.org), or at minimum reinstall the clean 5.2.4 from WordPress.org, delete any leftover updater/ directory, verify the files with "wp Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4 verify-checksums quick-pagepost-redirect-Quick Page/Post Redirect Plugin WordPress plugin before 5.2.4", clear the opcode cache, and treat any site that ran the tampered build as potentially compromised.

    Code execution Web app

    5.4 MEDIUM 0.1% exploit odds No fix listed
  27. CVE-2026-123847

    In JetBrains IntelliJ IDEA before 2026.3 command execution was possible via control characters in terminal command completion

    Code execution Jetbrains Intellij Idea

    5.3 MEDIUM 0.1% exploit odds Fix listed
  28. CVE-2026-102739

    ESPectre is a Wi-Fi motion sensing platform for ESP32. Prior to 3.0.0-rc1, Micro-ESPectre analysis tools, including tools/6_optimize_filter_params.py, tools/1_analyze_raw_data.py, and tools/10_train_ml_model.py, load NPZ datasets with NumPy pickle support enabled through micro-espectre/tools/csi_utils.py and a direct loader. A malicious dataset contributor or pull-request author can place an object-typed array in a .npz file under micro-espectre/data, causing pickle-controlled code to execute as the maintainer or developer when a documented analysis tool accesses the field. This issue is fixed in version 3.0.0-rc1.

    Code execution Device

    7.8 HIGH 0.1% exploit odds Fix listed
  29. CVE-2026-121750

    Node Version Manager (nvm) is a POSIX-compliant shell function for managing multiple node.js versions. In versions 0.40.5 through 0.40.8, `nvm_download_artifact` downloads a release archive directly to its final path in `$NVM_DIR/.cache` and verifies its checksum afterwards; on a checksum mismatch it reports an error but leaves the rejected archive in place. A later `nvm install --offline` of the same version installs that cached file without a checksum, so an archive that was just rejected gets installed and its `node` and `npm` run. This gives code execution to a party who can alter the downloaded tarball (for example a poisoned CDN or cache object) but cannot also forge the `SHASUMS256.txt` beside it, if the victim runs `nvm install --offline` for that version after the visible "Checksums do not match" failure and before any online retry. Version 0.41.0 only puts an archive in the cache once its checksum has passed.

    Code execution Library

    2.1 LOW 0.1% exploit odds Fix listed
  30. CVE-2026-21151

    Improper privilege management in TransitionPlayer prior to SMR Nov-2026 Release 1 allows local privileged attackers to execute arbitrary code with launcher privilege.

    Code execution Device

    8.3 HIGH 0.1% exploit odds No fix listed
  31. CVE-2026-123849

    In JetBrains IntelliJ IDEA before 2026.3 code execution was possible in untrusted projects via the Terraform File Watchers

    Code execution Jetbrains Intellij Idea

    7.0 HIGH 0.1% exploit odds Fix listed
  32. CVE-2026-57203

    Asterisk is an open source private branch exchange and telephony toolkit. Prior to Asterisk 20.20.1, 21.12.3, 22.10.1, 23.4.1 and Certified Asterisk 20.7-cert11, 22.8-cert3, `ast_loggrabber` executes a `Python script` from the world-writable `/tmp` directory. An attacker already authorized to open a shell session on the Asterisk server may place a malicious script there and gain elevated privileges or execute code when an administrator runs `ast_loggrabber` as a user with elevated privileges. This issue is fixed in Asterisk 20.20.1, 21.12.3, 22.10.1, 23.4.1 and Certified Asterisk 20.7-cert11, 22.8-cert3.

    Code execution

    6.9 MEDIUM 0.1% exploit odds Fix listed
  33. CVE-2026-86579

    Binary Planting Vulnerability in Hitachi Dynamic Link Manager for Windows. This issue affects Hitachi Dynamic Link Manager for Windows: from 5.9.5-00 before 9.0.4-01.

    Code execution OS & kernel

    8.2 HIGH 0.1% exploit odds No fix listed
  34. CVE-2026-120833

    labbench before 0.46.1 contains a deserialization of untrusted data vulnerability in read() and read_relational() that passes any file with a .p or .pickle extension to pandas.read_pickle based on the extension alone. Attackers who supply a crafted data file or relational data archive can execute arbitrary code when a user loads it with these functions.

    Code execution Library

    8.5 HIGH 0.1% exploit odds Fix listed
  35. CVE-2026-123308

    In JetBrains IntelliJ IDEA before 2026.3 code execution via the document import was possible in untrusted projects

    Code execution

    6.3 MEDIUM 0.1% exploit odds No fix listed
  36. CVE-2026-117526

    A vulnerability was found in reduxjs redux-devtools 5.0.0. Affected by this vulnerability is the function evalAction of the component Remotedev Server. The manipulation of the argument action results in code injection. The attack is only possible with local access. The exploit has been made public and could be used. The project was informed of the problem early through an issue report but has not responded yet.

    Code execution Library

    1.9 LOW 0.1% exploit odds No fix listed
  37. CVE-2026-111184

    NoMachine Named Pipe Incorrect Permission Assignment Local Privilege Escalation Vulnerability. This vulnerability allows local attackers to escalate privileges on affected installations of NoMachine. An attacker must first obtain the ability to execute low-privileged code on the target system in order to exploit this vulnerability. The specific flaw exists within the NoMachine Device Service. The issue results from incorrect permissions on a named pipe. An attacker can leverage this vulnerability to escalate privileges and execute arbitrary code in the context of SYSTEM. Was ZDI-CAN-31213.

    Code execution

    7.8 HIGH 0.1% exploit odds No fix listed
  38. CVE-2026-123311

    In JetBrains IntelliJ IDEA before 2026.3, 2026.2.4 command execution via a hidden run control in the Markdown preview was possible

    Code execution Jetbrains Intellij Idea

    5.3 MEDIUM 0.1% exploit odds Fix listed
  39. CVE-2026-15564

    Wildfly's IIOP listener may permit deserialization of input streams before authentication is performed. An attacker who can gain access to the IIOP port may be able to construct and send malicious objects and achieve deserialization, if the target EJB allows untyped method parameters, and depending on the potential gadget chains available in the running classpath.

    Code execution

    7.4 HIGH 0.1% exploit odds No fix listed
  40. CVE-2026-59081

    Edito CMS deserializes JSON data without further validation. A user (even an administrator) cannot exploit this vulnerability through the application's normal functionality. An attacker who could modify cache entries or inject them could execute code on the server. This issue was fixed in version 5.10.2607.  Hotfixes for older versions (5.8) have also been released.

    Code execution Web app

    8.4 HIGH 0.1% exploit odds Fix listed
  41. CVE-2026-95649

    IRISPowerscan insufficiently restrict the use of untrusted scripts. An authenticated user with access to script functionality can import or modify scripts that are executed by the application, potentially resulting in arbitrary code execution in the security context of the Windows user running IRISPowerscan.

    Code execution OS & kernel

    6.1 MEDIUM 0.1% exploit odds No fix listed
  42. CVE-2026-20798

    Uncontrolled search path for some Intel(R) Graphics Driver software installers within Ring 3: User Applications may allow an escalation of privilege. Unprivileged software adversary with an authenticated user combined with a high complexity attack may enable local code execution. This result may potentially occur via local access when attack requirements are present without special internal knowledge and requires active user interaction. The potential vulnerability may impact the confidentiality (high), integrity (high) and availability (high) of the vulnerable system, resulting in subsequent system confidentiality (none), integrity (none) and availability (none) impacts.

    Code execution OS & kernel

    5.4 MEDIUM 0.1% exploit odds No fix listed
  43. CVE-2026-119333

    draw.io Desktop is the official Electron build of draw.io. Prior to 31.5.2, the Windows application launched powershell and attrib by bare executable name from src/main/electron.js. When draw.io Desktop was not already running, opening a diagram from File Explorer made the diagram directory the current working directory, allowing a colocated attacker-controlled powershell.bat or attrib.exe to be resolved before the intended system executable. A victim who extracted an attacker-supplied archive and cold-opened its legitimate .drawio file could therefore execute attacker-controlled code when Local Fonts were enabled or when draft or backup attributes were updated. This issue is fixed in version 31.5.2.

    Code execution

    7.8 HIGH 0.1% exploit odds Fix listed
  44. CVE-2026-111573

    Incorrect permissions on the Checkmk Windows agent's data directory, caused by a race condition that discards the permission changes during service start, in Checkmk <2.5.0p18, <2.4.0p39, <2.3.0p52 and 2.2.0 (EOL) allow a local user to execute arbitrary code as SYSTEM by placing scripts in the agent's plugins or local folders, and to read the agent controller's private key.

    Code execution OS & kernel

    6.3 MEDIUM 0.1% exploit odds No fix listed
  45. CVE-2026-105477

    A vulnerability in Veeam Agent for Linux allows for local privilege escalation. Only installations that are managed by a Veeam Backup & Replication instance are affected.

    Code execution OS & kernel

    7.3 HIGH 0.1% exploit odds No fix listed
  46. CVE-2026-125396

    Ghostscript before 10.08.0 contains a -dSAFER bypass vulnerability in the .putgstringcopy procedure in gs_fonts.ps, which calls the privileged .forceput operator without checking operand types. Attackers can supply crafted PostScript or PDF documents that trigger a type error, recover .forceput via a custom error handler, and cause a use-after-free across save/restore to execute arbitrary code.

    Code execution

    8.5 HIGH 0.1% exploit odds Fix listed
  47. CVE-2026-114032

    The OpenSSL library bundled with the Illumio VEN is affected by an uncontrolled search path vulnerability. A low-privileged, authenticated local attacker could potentially leverage this vulnerability to execute arbitrary code with elevated privileges.

    Code execution Library

    7.8 HIGH 0.1% exploit odds No fix listed
  48. CVE-2026-108292

    IBM Process Mining 2.0.0 through 2.2.1 could allow an authenticated user to upload a file with dangerous types that could be executed by another user if opened.

    Code execution

    5.5 MEDIUM 0.1% exploit odds No fix listed
  49. CVE-2026-40573

    NovumOS is a custom 32-bit operating system written in Zig and x86 Assembly. Prior to version 0.25-beta.1, the `Nova` interpreter runs in Ring 0 with unrestricted kernel-memory access. Loading a malicious `.nv` script through the shell permits a local user to execute code with kernel privileges without security checks. The advisory identifies VM initialization in `zig/nova/vm.zig` as the affected component; script execution is not isolated in a restricted Ring 3 privilege domain. This issue is fixed in version 0.25-beta.1.

    Code execution OS & kernel

    8.6 HIGH 0.1% exploit odds Fix listed
  50. CVE-2026-121614

    Docker Sandboxes versions 0.31.0 before 0.50.0 do not neutralize repository-controlled Git configuration when running certain host-side Git commands. An attacker who can modify a host repository's Git configuration, for example through a direct-mode sandbox with the repository mounted read-write, can configure core.fsmonitor or another Git setting that invokes an external program. Git can execute the program during a later clone-mode fetch or, starting in version 0.39.0, an sbx kit push initiated by the user. The program runs on the host with the invoking user's permissions, outside the sandbox's microVM isolation. The read-only host repository mount provided to a clone-mode sandbox does not itself allow that sandbox to modify the host repository's Git configuration. The issue is fixed in version 0.50.0.

    Code execution Server

    7.1 HIGH 0.1% exploit odds Fix listed

CVE records published in 2026 (ids starting CVE-2026-), sorted by how likely they are to be exploited.