RCE
Monthly
Privilege-scoped remote code execution in SolarWinds Serv-U managed file transfer server stems from an insecure direct object reference that lets an authenticated domain administrator reference objects outside their intended authorization boundary to run code on the host. The flaw carries a 9.1 CVSS due to network reach, scope change, and full confidentiality/integrity/availability impact, though the practical bar is high because domain administrator access is a prerequisite, and impact is reduced on Windows deployments. No public exploit is identified at time of analysis and it is not listed in CISA KEV.
Privilege-to-root remote code execution in SolarWinds Serv-U managed file transfer server stems from an insecure direct object reference (CWE-639) that an authenticated domain administrator with read/write access to the home directory can abuse to execute code as root on the underlying host. The CVSS 3.1 score of 9.1 reflects a scope change from application-admin context to full OS compromise, with the impact explicitly noted as lower on Windows deployments. There is no public exploit identified at time of analysis and the issue is not listed in CISA KEV, so this is a high-severity but authentication-gated flaw rather than a mass-exploitable unauthenticated bug.
Remote code execution in SolarWinds Serv-U managed file transfer/FTP server allows an attacker to run arbitrary code remotely, executing as root on Unix/Linux deployments while impact is reduced on Windows. The flaw stems from an access-control weakness (CWE-284) and is corroborated by a vendor 'Authentication Bypass' tag, carrying a CVSS 9.1 (scope-changed) rating. No public exploit identified at time of analysis and it is not listed in CISA KEV, so exploitation is currently theoretical rather than confirmed.
Privilege escalation to remote code execution as root in SolarWinds Serv-U stems from an insecure direct object reference (IDOR) that lets an authenticated group administrator manipulate object references beyond their authority. The flaw carries a CVSS 9.1 rating because a scope change allows a lower-tier admin to fully compromise the underlying host on Linux/Unix, though impact is reduced on Windows deployments. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, but the high privilege prerequisite (group admin) is realistic in multi-tenant MFT environments.
Local privilege escalation in snap-confine - the SUID/set-capabilities helper snapd uses to build the confinement sandbox for snap applications - lets an unprivileged local user on Ubuntu 22.04, 24.04, and 26.04 LTS gain full root. The flaw only manifests where snap-confine is deployed with file capabilities (set-capabilities) instead of the traditional set-uid-root binary, where incorrect handling of ambient capabilities lets an attacker escape intended privilege boundaries and run arbitrary code as root. Reported by Canonical; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Root command injection in dracut's NetworkManager-based initrd network module allows an adjacent-network attacker to execute arbitrary code inside the initramfs during early boot. By answering a booting system's DHCP request with malicious root-path, next-server, or bootfile-name options, the attacker's payload is written unescaped into a temporary shell script and executed as root before the real OS starts. No public exploit is identified at time of analysis, and the flaw is not on CISA KEV, but a Red Hat advisory (RHSA-2026:26534) and upstream fix commit confirm the issue is real and patched.
Memory-corruption in Mozilla Firefox ESR 115.37 and 140.12 could allow attackers to run arbitrary code by triggering memory-safety bugs (CWE-119) that Mozilla observed showing evidence of memory corruption. Reported by Mozilla and tracked under advisories MFSA2026-69/70, the flaw is remotely reachable and carries a CVSS 9.8, though there is no public exploit identified at time of analysis and no active-exploitation (KEV) listing. Mozilla explicitly frames exploitation as presumptive ('with enough effort'), indicating no confirmed working exploit.
Memory-safety defects in Mozilla Firefox 152 and the Firefox ESR 115.37 and 140.12 release trains can corrupt process memory, and Mozilla states some showed evidence of memory corruption that could be leveraged to run arbitrary code. The flaws were resolved in Firefox 153, Firefox ESR 115.38, and Firefox ESR 140.13. There is no public exploit identified at time of analysis, and the EPSS probability is low (0.14%, 4th percentile), consistent with a batch of internally-found bugs rather than an actively exploited zero-day.
Memory corruption in Mozilla Firefox 152 and Firefox ESR 140.12 could allow attackers to run arbitrary code within the browser, per Mozilla advisories MFSA2026-68 and MFSA2026-70. The flaw is a batch of internally-discovered memory safety bugs (CWE-119) affecting the Firefox rendering/JavaScript engine, at least some of which showed evidence of exploitable memory corruption. No public exploit identified at time of analysis; the issues were reported by Mozilla itself and fixed in Firefox 153 and Firefox ESR 140.13.
Memory corruption in Mozilla Firefox 152 lets a remote attacker who lures a victim to a malicious web page trigger memory-safety violations that Mozilla assesses could be leveraged for arbitrary code execution. The flaw is one of a batch of internally-discovered memory safety bugs (CWE-119) rolled up under MFSA2026-68 and carries a CVSS 9.8 rating; there is no public exploit identified at time of analysis, and it is fixed in Firefox 153.
Remote code execution in Grav CMS 2.0.4 (fixed in 2.0.7) lets an authenticated user holding the admin.pages or api.pages.write permission plant a malicious Class::method callable directive in page frontmatter, which Blueprint::dynamicData() feeds unfiltered into call_user_func_array(). The injected command then executes as the web-server user for any subsequent visitor to the page, including unauthenticated ones. Reported by VulnCheck (CWE-94, CVSS 4.0 base 9.3); no public exploit identified at time of analysis and it is not listed in CISA KEV.
Stack-based buffer overflow in libssh's SFTP server allows a connected client to crash the server or potentially execute code by placing files with overlong names into a directory the server subsequently lists. The flaw lives in how libssh builds the SFTP 'longname' field, concatenating attacker-influenced filenames into a fixed-size stack buffer without adequate bounds checking. There is no public exploit identified at time of analysis and EPSS is very low (0.10%), but the memory-corruption nature (CWE-121) and libssh's broad deployment make it worth prioritized patching.
Arbitrary file upload in the MapSVG WordPress plugin (all versions through 8.14.0) lets authenticated Administrator-level users bypass file-type validation in the SVGFile constructor and upload arbitrary files, potentially achieving remote code execution on the host. The flaw stems from an incorrect conditional that skips validation entirely rather than enforcing it. Reported by Wordfence with a CVSS 3.1 base score of 7.2; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Arbitrary code execution in DayuanJiang next-ai-draw-io 0.4.13 arises from a path traversal flaw in the bundled MCP (Model Context Protocol) server, allowing an attacker who can influence a request to the mcp-server component to write or access files outside the intended directory and ultimately run code on the host. The affected product is a niche open-source AI-assisted draw.io diagramming application. No public exploit has been confirmed, though a third-party technical writeup exists; EPSS is low (0.43%, 35th percentile) and the flaw is not in CISA KEV.
Cross-site scripting in DayuanJiang next-ai-draw-io 0.4.13 lets a remote unauthenticated attacker inject arbitrary script through the mcp parameter, executing in the victim's browser when they open a crafted link. The 'arbitrary code' claim and RCE tag in the source data most plausibly refer to arbitrary JavaScript in the browser (CWE-79 reflected XSS) rather than server-side code execution. EPSS is low (0.29%, 21st percentile), there is no CISA KEV listing, and no public exploit is confirmed, though third-party GitHub writeups referencing this CVE exist.
Remote arbitrary code execution in Bytebot (bytebot-ai self-hosted AI desktop agent, commit 3d37894 dated 2025-09-11) allows attackers to run OS commands by supplying a crafted path to the agent's `computer_write_file` action. The flaw is classed as OS command injection (CWE-78) and carries a CVSS 9.8, but real-world exploitation probability is currently low (EPSS 0.17%, 7th percentile) and it is not on CISA KEV. A public gist referenced by the CVE appears to document the issue and may contain proof-of-concept details, though POC status is not explicitly confirmed in the source data.
Privilege escalation to remote code execution in dotCMS 21.02 through 26.06.22-03 lets a low-privileged authenticated backend user abuse improper authorization checks in the ToolGroupResource (REST) and RoleAjax (DWR) endpoints to self-assign the administrative layout and self-grant the CMS Administrator role, then upload a crafted OSGi bundle whose BundleActivator runs arbitrary shell commands on the host. The endpoints failed to verify the caller was an admin before honoring layout- and role-assignment requests, so any backend account could bootstrap itself to full administrator. There is no public exploit identified at time of analysis, but an upstream fix (PR #36344) is available and the vendor-supplied CVSS 4.0 score is 9.4 (Critical).
Arbitrary code execution in Adobe DNG SDK version 1.7.1 (build 2536) and earlier arises from a stack-based buffer overflow triggered when parsing a maliciously crafted image file, allowing an attacker to run code in the context of the current user. Applications that embed the DNG SDK to decode Digital Negative (DNG) raw camera files are affected, and successful exploitation requires the victim to open the malicious file. No public exploit identified at time of analysis, and the flaw is not listed in CISA KEV.
Authenticated PHP code execution in ThemeXpert's Quix Page Builder Pro extension for Joomla lets a privileged builder user (holding core.create or core.edit permissions) embed raw PHP tags in element content that are subsequently executed when the server's view-cache layer include()s the cached template. Exploitation requires caching to be enabled, which is the default, and yields full remote code execution on the underlying web server. No public exploit identified at time of analysis, though the reporter is Joomla itself and the CVSS 4.0 base score is 8.9 (High).
Remote code execution in Chamilo LMS 1.11.40 and earlier allows any user enrolled in a course to run arbitrary PHP through the main/inc/ajax/lang.ajax.php endpoint, which is guarded only by api_protect_course_script(true) and is therefore reachable by students, teachers, and DRH (HR) roles rather than administrators. The root cause is an improper-authorization defect (CWE-285) that exposes a dangerous language-editing endpoint to low-privilege course members. No public exploit code is identified at time of analysis and it is not listed in CISA KEV, but a vendor security advisory (GHSA-4hwq-pv7c-3928) has been published.
Stack-based buffer overflow in rsyslog's optional mmpstrucdata module allows a remote, unauthenticated attacker to crash the rsyslog process and interrupt log collection, with theoretical code execution on weakly hardened builds. Affected versions span rsyslog 7.5.4 through 8.2605.x; the fix is included in 8.2606.0. No public exploit code or active exploitation has been confirmed, but the crash was independently reproduced over both TCP and UDP on Debian 12 with rsyslog 8.2302.0, and the corrective PR (#6991) was merged publicly before coordinated disclosure completed.
Remote code execution in Piwigo 16.3.0 and earlier lets an unauthenticated attacker inject arbitrary PHP through the web installer, which writes unsanitized POST parameters (prefix, dbpasswd, dbhost, dbname, dbuser) into local/config/database.inc.php where they execute on every page load. The bug is a PHP 8 regression: the addslashes() guard is gated on get_magic_quotes_gpc(), removed in PHP 8.0, so escaping is silently skipped and raw input reaches PHP source. No public exploit identified at time of analysis, but the flaw is trivially weaponizable when the installer is reachable, yielding full server compromise.
Arbitrary code execution in ProFTPD's mod_sftp module (versions before 1.3.9c and 1.3.10rc3) allows authenticated low-privilege SFTP users to run code in the server process by abusing SFTP packet reassembly in fxp.c. The flaw stems from an incorrectly conditioned reallocation of the 16 KB reassembly buffer that corrupts memory-pool freelist metadata, ultimately hijacking pr_fsio_stat() to reach system(). Reported by VulnCheck with a vendor patch available; no public exploit identified at time of analysis, though the advisory describes a complete exploitation primitive.
Remote code execution in Apache Syncope's connector subsystem allows an administrator holding sufficient entitlements to run arbitrary Groovy scripts through scripted REST and SQL connectors, escaping the intended sandbox and executing code on the host. It affects Syncope 3.0.0-M0 through 3.0.16, 4.0.0-M0 through 4.0.6, and 4.1.0-M0 through 4.1.1, and is rooted in improper isolation (CWE-653) of the scripting engine. Although the published CVSS is 9.8, the vendor description states an authenticated privileged administrator is required, and there is no public exploit identified at time of analysis.
Local privilege escalation to root in the QEMU Guest Agent (qga) affects Red Hat Enterprise Linux 6 through 10, RHEL for NVIDIA 26, and OpenShift Container Platform 4, where a local unprivileged guest user abuses the guest-ssh-add-authorized-keys command handler through symlink manipulation to seize ownership of arbitrary root-owned files. Two attack variants exist - a deterministic directory-symlink bypass and a TOCTOU file-symlink race - both requiring an external management layer such as libvirt to invoke the code path. No public exploit is identified at time of analysis and it is not listed in CISA KEV, but the CVSS 7.3 rating and clean path to full root access make it a meaningful local escalation risk in multi-tenant virtualization estates.
Path traversal in Menyoo (MenyooSP), a popular GTA V single-player trainer/mod, before commit 729aa48 lets a local user supply crafted file names (containing traversal sequences) to the Spooner, VehicleSpawner, WeaponOptions, and PedComponentChanger save/create-folder/rename functions, writing or renaming files outside the intended data directories. Because the resulting XML/config files can be planted in attacker-chosen locations, this escalates to arbitrary code execution in the victim's context. There is no public exploit identified at time of analysis and it is not in CISA KEV; the fix (added IsSafePath() input validation) is publicly visible in PR #550.
Arbitrary code execution in SJCAM SJ4000-Air action cameras (firmware V1.4C and earlier), and unbranded whitelabel devices built on the same v1.4C firmware, is possible via a maliciously crafted FEX file - the Allwinner firmware/partition configuration format used by these AllwinnerTech-based devices. The flaw is classified as code injection (CWE-94) with a supplied CVSS of 9.8; no public exploit or CISA KEV listing exists at time of analysis, though a detailed reverse-engineering writeup of the device firmware is publicly referenced. The AV:N network rating in the supplied vector is inconsistent with the description's file-processing mechanism and warrants verification.
Poisoned-pipeline (pwn request) code execution in the Meshtastic firmware repository's GitHub Actions CI allows any external user to run arbitrary code inside the project's trusted CI context prior to firmware 2.7.21.1370b23. Because main_matrix.yml (and related workflows) triggered on pull_request_target while the check, build, and build-debian-src jobs checked out and executed code from the attacker's fork with no approval gate, a pull request from an author_association of NONE automatically ran attacker-controlled files with access to repository secrets and an elevated GITHUB_TOKEN. This is CWE-94 code injection scoring CVSS 10.0; no public exploit identified at time of analysis, though a linked Google Drive writeup may contain proof-of-concept detail.
Arbitrary code execution in keras-team/keras 3.15.0 arises because the public keras.layers.TorchModuleWrapper.from_config method calls torch.load(..., weights_only=False) without any explicit unsafe opt-in, so a maliciously crafted Keras layer configuration deserialized outside a SafeModeScope(True) context triggers execution of attacker-controlled PyTorch pickle payloads (CWE-502). Any application or pipeline that loads untrusted or third-party Keras model/layer configs is affected; CVSS is 7.8 (High) with a local vector requiring user interaction. There is no public exploit identified at time of analysis and the flaw is not listed in CISA KEV, but the vulnerability class is well understood and reliably weaponizable via standard pickle reduce gadgets.
Session hijacking and keystroke injection in the urwid Python library's web display backend (urwid/display/web.py) let attackers take over active web terminal sessions and reach OS-level code execution with the session owner's privileges. The urwid_id session token is generated with Python's non-cryptographic Mersenne Twister PRNG, so a remote attacker who observes roughly 334 session IDs (e.g. via the X-Urwid-ID response header) can reconstruct PRNG state and predict every past and future token, while any local user can read the same token from world-listable FIFO filenames in /tmp. No public exploit has been identified at time of analysis; the issue was reported by VulnCheck and carries a 9.2 (CVSS 4.0) severity.
Arbitrary Python code execution in scikit-hep uproot (uproot5) arises because the library dynamically generates and compiles reader classes from a ROOT file's TStreamerInfo records, interpolating file-controlled streamer metadata (such as streamer element names) into generated Python source without repr()/!r safe quoting. A crafted ROOT file can inject expression-breaking content that is evaluated in the process opening the file, giving code execution to whoever processes attacker-supplied ROOT data. Reported by VulnCheck; no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Remote code execution in VMware Avi Load Balancer exposes organizations running versions from 22.1.1 through 32.1.1 to arbitrary code injection by low-privileged authenticated users with network access. The flaw is classified as CWE-94 code injection, meaning attacker-supplied input reaches an execution context without adequate sanitization. Broadcom has released patches across all affected branches; no public exploit code has been identified at time of analysis, and the EPSS score of 0.68% (48th percentile) reflects modest current exploitation probability despite the high CVSS 8.8 rating.
Remote code execution in VMware Avi Load Balancer exposes the Avi Control Plane to arbitrary code execution by any low-privileged, network-connected user, stemming from a code injection flaw (CWE-94) in the controller. Multiple major version branches from 22.1.1 through 32.1.1 are confirmed affected, with vendor-released patches available across all branches. No public exploit code or CISA KEV listing has been identified at time of analysis, though the CVSS 8.8 rating and full high-impact triad (C:H/I:H/A:H) make this a high-priority patch target for any deployment where the control plane is network-reachable.
Remote code execution and information disclosure in IBM Db2 Genius Hub (1.1, 1.1.1, 1.1.2) and IBM Agentics 1.0 stem from the use of inherently dangerous functions invoked without sufficient input restrictions, allowing remote attackers to run arbitrary code or read sensitive data. IBM's PSIRT reported the flaw with a 9.8 critical CVSS reflecting network-reachable, unauthenticated exploitation with full confidentiality, integrity and availability impact. There is no public exploit identified at time of analysis, and the low EPSS score (0.33%, 25th percentile) suggests exploitation is not yet widespread.
Heap-based buffer overflow in IBM Storage Protect Client (versions 8.1.0.0-8.1.27.1 and 8.2.0.0-8.2.1.0) lets a remote attacker corrupt heap memory through improper bounds checking, potentially achieving arbitrary code execution or crashing the affected process. The flaw carries a CVSS 3.1 base score of 9.8, but no public exploit identified at time of analysis and the EPSS probability is low (0.46%, 37th percentile). Note that the affected component is the backup Client, while the description references crashing 'the server,' introducing ambiguity about the true exploitation path.
Unauthenticated remote code execution in IBM Langflow OSS versions 1.0.0 through 1.10.1 lets anonymous attackers run arbitrary code by abusing the public flow build endpoint (/api/v1/build_public_tmp/{flow_id}/flow). The CVSS 3.1 base score is 9.8 (network, low complexity, no privileges or user interaction). No public exploit is identified at time of analysis, and the EPSS probability is modest at 0.44% (36th percentile), indicating exploitation is technically trivial but not yet observed at scale.
Extension-denylist bypass in the Flask-Reuploaded (flask_uploads) Python library allows an attacker who controls the name override passed to UploadSet.save() to store dangerous script files by varying letter case (e.g. shell.PHP, evil.pHp), defeating the very script-blocking denylist the library documents. This is an incomplete-fix variant of CVE-2026-27641: the v1.5.0 fix re-validated overridden names through the case-preserving extension() helper instead of the case-normalizing lowercase_ext() used by the default path, so uppercase extensions slip past a lowercase-tokened denylist. Publicly available exploit code exists (a self-contained PoC verified against Flask-Reuploaded==1.5.0), though there is no evidence of active exploitation and no EPSS score was provided.
Arbitrary code execution in Adobe Acrobat Reader arises from a heap-based buffer overflow (CWE-122) triggered when a victim opens a maliciously crafted file, allowing an attacker to run code in the context of the current user. All affected versions covered by Adobe advisory APSB26-63 are impacted, and successful exploitation grants full high impact to confidentiality, integrity, and availability. There is no public exploit identified at time of analysis, and the issue is not listed in CISA KEV, so risk is currently latent rather than active.
Remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows an authenticated remote attacker to run arbitrary code because validation is not fully enforced on Model Context Protocol (MCP) server configuration files, permitting injection of attacker-controlled commands or server definitions. IBM has published a fix advisory (node/7278932). With CVSS 8.8 (PR:L), a low-privileged user of the Langflow instance can escalate to full code execution on the host; there is no public exploit identified at time of analysis and it is not listed in CISA KEV.
Remote code execution in WordPress 6.9.x (before 6.9.5) and 7.0.x (before 7.0.2) arises from a REST API batch endpoint route confusion flaw that, when chained with the author__not_in WP_Query SQL injection (CVE-2026-60137), lets attackers inject SQL and ultimately execute arbitrary code on the underlying host. Publicly available exploit code exists (a ProjectDiscovery Nuclei template and WPScan advisory), and the CISA SSVC framework rates the flaw as automatable with total technical impact, though it is not currently listed in CISA KEV. The vulnerability was reported by WPScan and patched in the WordPress 7.0.2 and 6.9.5 releases.
Remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows authenticated users to override arbitrary component parameters at runtime through the API, bypassing the parameter filtering logic in the apply_tweaks() function (CWE-94 code injection). Because a low-privileged authenticated account is sufficient and the impact is full compromise of confidentiality, integrity, and availability (CVSS 8.8), this maps to server-side code execution on the Langflow host. There is no public exploit identified at time of analysis, and the vulnerability is not listed in CISA KEV.
Remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows attackers to run arbitrary code by planting a malicious pickle payload that the AsyncDiskCache deserializes with unauthenticated pickle.loads(). Any actor able to influence cached data - via file system access, crafted workflow inputs, custom components, or API manipulation - gains full compromise at the privilege of the Langflow server process. Rated CVSS 9.9 and reported by IBM with a vendor patch available; no public exploit identified at time of analysis.
Authenticated remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 lets any logged-in user run arbitrary Python — and therefore arbitrary OS commands — with the privileges of the Langflow server process. The flaw is a design failure in the POST /api/v1/validate/code endpoint, which passes user-supplied code straight into Python's exec() with no sandbox, validation, or privilege drop. Rated CVSS 9.9 by IBM; no public exploit identified at time of analysis, though the trivial exec()-based trigger makes exploitation straightforward once authenticated.
Unauthenticated remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 stems from broken webhook authentication that skips API key validation whenever WEBHOOK_AUTH_ENABLE is False - which is the shipped default. Any remote attacker who learns a flow's UUID can invoke that flow's webhook endpoint as though they were its owner, running arbitrary flow logic and reaching RCE. No public exploit identified at time of analysis, but the CVSS 9.8 rating and default-vulnerable configuration make this a high-priority patch.
Privilege escalation to remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows an authenticated user to elevate their account to superuser by directly manipulating the underlying database, then execute arbitrary system commands and achieve full compromise of the host running the Langflow service. The flaw (CWE-94, code injection) carries a critical CVSS 9.9 with a scope change, meaning code runs with the Langflow service account's permissions beyond the vulnerable component. No public exploit identified at time of analysis, and it is not listed in CISA KEV, but a vendor patch is available.
Arbitrary Python code execution in IBM Langflow OSS 1.0.0 through 1.10.0 (Langflow versions up to 1.9.2) allows authenticated users with flow-creation privileges to bypass the allow_custom_components=false control via the Policies component's ToolGuard integration. Because validation only inspects the main component code and ignores dynamic CodeInput fields, attackers persist malicious Python in a flow that runs server-side when a guarded tool executes. No public exploit has been identified at time of analysis, but the flaw is IBM-confirmed with a vendor patch and carries a critical 9.9 CVSS due to cross-tenant escalation through the agentic MCP update_flow_component_field tool.
Unauthenticated remote code execution in IBM Langflow OSS versions 1.0.0 through 1.10.0 lets any network-reachable attacker achieve full RCE on default deployments by chaining two endpoints: /api/v1/auto_login, which mints SUPERUSER tokens to any caller, and /api/v1/validate/code, which passes user-supplied code to Python's exec(). CVSS is 9.8 (AV:N/AC:L/PR:N/UI:N) and a vendor patch is available; there is no public exploit identified at time of analysis, and the flaw is not listed in CISA KEV.
Code injection leading to remote code execution affects IBM Db2 versions 11.5.0 through 11.5.9 and 12.1.0 through 12.1.4 when an attacker can influence the JDBC connection URL passed to the driver. An attacker who controls the JDBC URL (typically via an application that builds connection strings from user input) can inject malicious properties or references that cause arbitrary code to execute in the context of the connecting application/driver host. IBM has released a fix; there is no public exploit identified at time of analysis and the issue is not listed in CISA KEV.
Arbitrary code execution in ForgeCode (tailcallhq/forgecode), an AI pair-programming CLI, occurs because the tool auto-loads and honors a repository's project-local .mcp.json at startup without asking the user. A malicious repository can define mcpServers entries with arbitrary command/args (e.g. bash -c), so simply running the forge CLI inside a cloned untrusted repo spawns attacker-chosen commands as the invoking developer. Reported by VulnCheck with a vendor patch available; no public exploit tool is identified at time of analysis, though the CVE description itself embeds a working payload example.
Unauthenticated SQL injection in Sangoma Switchvox SMB Edition 8.3 (build 104997) lets remote attackers execute arbitrary SQL against the backend PostgreSQL database via the /pa endpoint, which concatenates the attacker-supplied PhoneIP value from a <PolycomIPPhone> XML body into queries without parameterization. Because SQL access on this PostgreSQL deployment can be escalated to command execution, a single crafted, unauthenticated request yields full database compromise and remote code execution on the VoIP appliance. There is no public exploit identified at time of analysis, though vendor SRA published a technical writeup, and the CVSS 4.0 base score is 9.3 (Critical).
Arbitrary file write in GitHub Enterprise Server (all releases before 3.22) lets an attacker who already has code execution inside the Dependabot updater container escape the intended dependency-file directory via path traversal and symlink manipulation, planting attacker-controlled files anywhere in a repository - including GitHub Actions workflows under .github/workflows/. Because path validation checked the declared rather than the effective (symlink-resolved) path, an injected workflow can run with the repository's Actions secrets when the repo uses pull_request_target or auto-merge. There is no public exploit identified at time of analysis and it is not listed in CISA KEV; CVSS 4.0 base score is 8.6 (High).
Arbitrary code execution in Cursor for Windows 3.2.16 lets remote attackers run code on a developer's machine by planting a malicious git.exe in a repository's root directory. Because Cursor resolves and executes the workspace-resident git.exe at IDE startup and on a recurring timer, simply cloning and opening a crafted repository triggers execution under the current user's privileges with no explicit user action beyond opening the project. Publicly available exploit code exists and the flaw was reported by VulnCheck; there is no CISA KEV listing or active-exploitation confirmation in the data.
Remote code execution is possible in the ProfilePress WordPress plugin (Paid Membership, Ecommerce, User Registration) in all versions through 4.16.18, where the DigitalProducts UploadHandler unconditionally registers an upload_mimes filter that whitelists executable extensions (.exe, .apk, .msi) across every WordPress upload context site-wide. Authenticated attackers with author-level access or above (per CVSS PR:L) can leverage this expanded MIME allowlist to upload executable files, leading to code execution. There is no public exploit identified at time of analysis, and this CVE is not listed in CISA KEV.
Remote code execution in the Generic OEM UZ801_v2.1 4G LTE Router firmware V3.4.3 allows an unauthenticated remote attacker to run arbitrary commands through the exposed sbin/adbd (Android Debug Bridge daemon) component. The CVSS 3.1 base score is 9.1 (AV:N/AC:L/PR:N/UI:N) with high confidentiality and integrity impact, reflecting full command execution on the device. No CISA KEV listing is present and EPSS is low (0.35%, 27th percentile), but a researcher GitHub repository is referenced, indicating publicly available exploit code likely exists.
Hardcoded default administrator password in Pheditor (all versions prior to 2.0.6) lets remote unauthenticated attackers log in as admin using the string 'admin' — the SHA-512 hash of which is baked into pheditor.php line 11 — and then abuse the built-in terminal and file-upload features for immediate remote code execution. Any deployment left on defaults is trivially compromised because there is no forced password change, expiry, lockout, or setup wizard. A working PoC is published in the GitHub Security Advisory, though there is no public exploit identified in active in-the-wild campaigns and no EPSS/KEV data is provided.
Build-time remote code execution in Nuclio (serverless/FaaS platform) versions <= 1.15.27 lets remote attackers run arbitrary OS commands as root inside the function-builder container. The Java runtime renders user-supplied runtimeAttributes.repositories values into a Groovy build.gradle file via Go's text/template using the non-escaping {{ . }} action, so an attacker can inject a closing brace to break out of the repositories {} block and append Groovy that Gradle executes during its configuration phase. Because the Dashboard API defaults to NOP (no) authentication, exploitation needs no credentials; a dynamically-verified proof-of-concept is published in the vendor advisory, though there is no public exploit identified as being used in active attacks.
Remote kernel heap corruption in the illumos SCTP stack lets an unauthenticated attacker send a crafted INIT ACK packet with malformed address parameters to trigger an out-of-bounds access during packet classification, potentially leading to remote code execution or a kernel panic. The flaw affects illumos-gate and downstream distributions (OmniOS, SmartOS/Triton) built before commit 53a3efde, and has existed since 2010 (commit a5407c02). There is no public exploit identified at time of analysis and the CVSS 4.0 exploit-maturity metric is Unreported (E:U), but the attack requires no authentication or user interaction.
Arbitrary file disclosure in Envoy Gateway's controller allows a low-privileged tenant who can create an EnvoyExtensionPolicy to read sensitive files from the gateway controller pod. A path-normalization gap in the security.lua critical-path check (CWE-20) lets attacker-supplied Lua bypass the read restriction using double-slash paths (e.g. //etc/passwd), exposing Kubernetes service-account tokens, TLS certificates, and process environment. No public exploit identified at time of analysis, but the vendor advisory documents the exact bypass and CVSS is rated 9.1.
Code injection in remorses/genql (the @genql/cli TypeScript GraphQL client generator) before 6.3.4 lets an actor who controls the GraphQL schema fed to genql smuggle arbitrary TypeScript/JavaScript into the generated schema.ts file, which then executes when a downstream consumer bundles and imports the client. The flaw stems from GraphQL type/field descriptions containing a `*/` sequence that breaks out of the generated JSDoc comment block. There is no public exploit identified at time of analysis and it is not in CISA KEV; the issue was responsibly disclosed (issue #197) and reported by CISA (cisa-cg), with a vendor patch in 6.3.4.
Code injection in Frogman before 1.6.2 lets an authenticated PERM_WRITE caller of the MCP/HTTP dialplan API inject arbitrary Asterisk directives - including System() and Set(SHELL(...)) - into extensions_custom.conf, yielding remote command execution on the PBX host. The flaw stems from fm_dialplan_apply sanitizing only the context name while passing template parameters such as greeting, dest, url, extension, code, and file into generated dialplan text unchecked. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV, but the CVSS is 9.9 and the fix (v1.6.2) plus root-cause commit are public.
Local privilege escalation to arbitrary code execution affects Lenovo App Store, a software-distribution client shipped exclusively on Lenovo devices in the Chinese market. A local authenticated user who can interact with the application can leverage its excessive privileges (CWE-250) to run arbitrary code at an elevated privilege level, with high confidentiality, integrity, and availability impact on the host. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, so no active exploitation is confirmed.
Local code execution in Lenovo App Store (Chinese-market distribution only) lets an authenticated local user abuse a path traversal weakness to write or load files outside intended directories and run arbitrary code. The flaw was self-reported by Lenovo and carries a CVSS 4.0 base score of 7.0; there is no public exploit identified at time of analysis and it is not listed in CISA KEV. Impact is high to confidentiality, integrity, and availability of the local system, but exploitation requires existing low-privilege local access plus user interaction.
Local code execution in Lenovo Legion Zone and the Lenovo App Store (China-market Windows builds) allows an authenticated local user to run arbitrary code when either application is installed on a non-system partition. The flaw stems from insecure directory/file permissions (CWE-277) that a low-privileged user can abuse to plant or modify executable content later run with higher privilege. No public exploit is identified at time of analysis and it is not listed in CISA KEV; impact is confined to systems where these Chinese-market Lenovo utilities are deployed to a non-default drive.
Arbitrary JavaScript execution in Microsoft's @prompty/core TypeScript loader (2.0.0-alpha.1 through 2.0.0-beta.2) allows an attacker who supplies a crafted .prompty file to run code during prompt loading. The loader passed .prompty markdown to gray-matter without disabling its executable 'js'/'javascript' frontmatter engines, so a '---js' frontmatter block is evaluated as Node.js at parse time. Exploitation requires the victim's application to load the malicious file (UI:P); no public exploit is identified at time of analysis and it is not on CISA KEV, but the fix commit ships a working proof-of-concept test.
Unauthenticated remote code execution in SGLang (versions 0 through 0.5.14) arises when the expert-parallel backup subsystem binds a ZeroMQ PULL socket to a routable interface without authentication or safe deserialization, letting a network attacker send a crafted pickle payload that executes arbitrary code in the serving process. It affects deployments where the elastic expert-parallel backup feature is enabled and the socket is reachable. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Command injection in Microsoft Kiota before 1.32.5 lets a malicious or compromised OpenAPI description dictate the install command that Kiota presents to developers. When a developer runs `kiota info` (or the VS Code extension's `kiota info --json` dependency-install flow) against an attacker-controlled spec, Kiota reads the `x-ms-kiota-info.languagesInformation.<language>.dependencyInstallCommand` field plus attacker-chosen dependency name/version values and surfaces them as its trusted recommended install command, achieving arbitrary command execution when that command is run. No public exploit has been identified at time of analysis and it is not in CISA KEV, but the fix explicitly removes the untrusted extension field.
PHP code generation in Microsoft Kiota prior to 1.32.4 permits injection of arbitrary PHP code when the generator processes attacker-controlled OpenAPI specification fields, because the shared `SanitizeDoubleQuote()` helper in `Writers/StringExtensions.cs` fails to escape the `$` character before embedding schema-derived strings into PHP double-quoted literals. Developers or CI/CD pipelines that generate PHP API clients from untrusted or third-party specifications will produce PHP files containing live interpolation expressions (`${...}`, `$var`, `{$obj->prop}`) that execute attacker-chosen code at PHP runtime in production environments. No public exploit has been identified and SSVC rates exploitation as currently none, but technical impact is rated total; the realistic risk is tightly gated on whether teams consume external or attacker-influenced specs without generated-code review.
Code injection in Microsoft Kiota's Python client generator (versions prior to 1.32.0) allows an attacker who controls an OpenAPI description to embed arbitrary Python that executes when a developer generates and imports the resulting client. The flaw stems from unsanitized x-ms-enum.values[].description strings being emitted as inline comments; embedded newlines break out of the comment and run at module scope. No public exploit identified at time of analysis, and the issue is not in CISA KEV. The assigned CVSS 3.1 base score is 7.5, though its impact metrics (C:N/I:N/A:H) understate what is functionally full code execution.
Code injection in Microsoft Kiota Ruby code generator allows remote code execution through crafted OpenAPI specifications. Kiota versions prior to 1.32.0 fail to escape '#' characters in schema-derived strings embedded in Ruby double-quoted literals, enabling Ruby interpolation sequences such as '#{code}' to execute arbitrary commands when the generated client code runs. While no active exploitation is confirmed, the attack is automatable, EPSS indicates moderate probability (1.47%), and the vendor has released a fix.
Code-generation injection in Microsoft Kiota's C# XML documentation-comment writer allows an attacker who controls or compromises an OpenAPI specification to inject arbitrary C# statements into generated client source files. Affected packages Microsoft.OpenApi.Kiota and Microsoft.OpenApi.Kiota.Builder in versions prior to 1.29.1 and in the 1.30.0-1.32.2 range fail to strip C# line-terminator characters from description and externalDocs fields before emitting them as single-line /// doc comments, enabling payload breakout into compilable namespace-level code. No public exploit confirmed in the wild at time of analysis, though SSVC reports a POC exists and rates technical impact as total; exploitation culminates when a developer or CI pipeline builds the tainted generated output.
Sandbox escape in PipeWire's PulseAudio compatibility layer lets a low-privileged process inside a confined environment such as Flatpak load an attacker-controlled library and execute arbitrary code on the host user's session. The flaw (CWE-427) is present in PipeWire as shipped in Red Hat Enterprise Linux 7, 8, 9 and 10 and carries a scope-changing CVSS 3.1 score of 8.8. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Root-level remote code execution in Canonical's ubuntu-pro-client (formerly ubuntu-advantage-tools) allows an attacker who can spoof or tamper with the Ubuntu Pro contract server response to inject arbitrary APT source lines and install malicious root-owned packages. The client builds /etc/apt/sources.list.d entries from the server's directives.suites[] and directives.aptURL fields via Python str.format() without escaping or newline filtering, and passes additionalPackages[] positionally into a root-run apt-get install. Because the component is preinstalled on supported Ubuntu Server releases and auto-attaches on cloud Ubuntu Pro images the exposure is broad; there is no public exploit identified at time of analysis and it is not listed in CISA KEV.
Arbitrary file deletion in the Uncanny Automator WordPress plugin (all versions through 7.3.1.4) lets unauthenticated attackers delete any file on the server via untrusted PHP object deserialization in the fr_token function, and deleting a critical file such as wp-config.php can pivute WordPress into an installation/setup state that yields remote code execution. Exploitation is gated by a specific setup: a Forminator form must be connected to an Uncanny Automator recipe whose trigger is configured for 'Everyone', which is what exposes the deserialization sink to anonymous form submissions. A self-contained gadget chain via Action_Helpers_Email::__destruct() ships inside the plugin, so no external gadget library is needed; reported by Wordfence with no public exploit identified at time of analysis and no CISA KEV listing.
Missing authentication in @andrea9293/mcp-documentation-server v1.13.0 exposes its document-management Web UI/API on all network interfaces (0.0.0.0:3080) instead of localhost, letting any adjacent-network client read, add, search, and delete documents in the MCP knowledge base without credentials. Rated CVSS 8.8 (CWE-306), the flaw is exploitable by unauthenticated attackers reachable over the same LAN, VM network, Docker bridge, or VPN. A detailed proof-of-concept exists in the reporter's advisory, though there is no public exploit identified as being used in active attacks and it is not listed in CISA KEV; the vendor fixed it in v1.13.1.
Sandbox escape with SYSTEM-level code execution in Sandboxie-Plus before 1.17.6 allows a process confined inside a sandbox to run arbitrary code in an unsandboxed host process. The privileged GuiServer service trusts attacker-supplied thread and function-pointer values from a GUI window-hook registration request and dispatches them via QueueUserAPC while running as SYSTEM, defeating the core isolation guarantee of the product. No public exploit identified at time of analysis and the flaw is not listed in CISA KEV, but the mechanism is a full sandbox-to-host escape.
Authenticated remote code execution in 9Router before 0.5.2 lets a logged-in attacker run arbitrary OS commands on the host by chaining a Host-header spoof (to reach routes that are supposed to be localhost-only) with unsanitized MCP plugin arguments that flow into child_process.spawn() via the /api/mcp//sse endpoint. Rated CVSS 8.8 (CWE-78, OS command injection), it is fixed in 0.5.2; no public exploit or CISA KEV listing exists at time of analysis, though a vendor advisory and fix commit are published.
Remote authenticated code execution in DataEase before 2.10.23 lets a low-privileged user abuse the Excel/datasource upload workflow to run arbitrary Java. By uploading a crafted payload.zip via /datasource/upload and defining an H2 datasource that references the archive through the zip: protocol, an attacker forces CalciteProvider.jdbcFetchResultField to execute a query that triggers precompiled Java aliases embedded in test.mv.db. The vendor CVSS 4.0 vector scores this 7.1 (confidentiality-only), but the underlying primitive is full arbitrary code execution; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Reflection-based remote code execution in DataEase before 2.10.23 lets an authenticated operator who can configure a Redshift datasource achieve arbitrary code execution on the server. The Redshift JDBC driver loads an attacker-controlled rsjdbc.ini from the system temp directory and honors a malicious socketFactory pointing at Spring's FileSystemXmlApplicationContext, turning a normal JDBC connection into a code-execution chain. No public exploit is identified at time of analysis and it is not in CISA KEV, but the vendor has confirmed and patched the flaw in 2.10.23.
Server-side arbitrary code execution in TDengine time-series database (versions prior to 3.4.1.15) allows a database user holding the 'create udf' privilege to upload a malicious shared library, register it as a user-defined function (e.g. named 'eval'), and execute attacker-controlled native C code on the server by invoking that function through ordinary SQL queries. This is a privileged-user code injection issue (CWE-94) with high confidentiality, integrity, and availability impact; no public exploit identified at time of analysis and it is not listed in CISA KEV. The flaw is fixed in version 3.4.1.15.
Stack-based buffer overflow in TDengine (open-source IoT time-series database) versions 3.4.1.6 and earlier allows an authenticated user to trigger a one-byte out-of-bounds write by submitting SQL containing crafted escape sequences (\%, \_, or \x), leading to denial of service and potentially remote code execution. The flaw sits in the SQL parser's trimString() routine, which validates only a single byte of remaining space in the tmpTokenBuf stack buffer before writing escape-processed data. There is no public exploit identified at time of analysis, and the issue is not listed in CISA KEV; it is fixed in version 3.4.1.14.
Remote code execution in Cherry Studio 1.2.2 through 1.9.12 lets attackers who control search-provider content run arbitrary Node.js code because the SearchService loads that content into an Electron BrowserWindow with nodeIntegration enabled and contextIsolation disabled. An attacker controlling a search engine provider, search result page, or provider settings page can execute JavaScript with full Node.js privileges under the OS account running the app. Publicly available exploit code exists (VulnCheck/Mundi-Xu gist) and a vendor patch (commit 1518530) is available; the flaw is not currently listed in CISA KEV.
Authenticated remote code execution in LangBot (pip package 'langbot', versions <= 4.10.5) allows any logged-in user to run arbitrary OS commands on the host by registering a malicious 'STDIO' MCP server through the Extensions > MCP configuration UI. Because LangBot passes the user-supplied command straight to the MCP SDK's StdioServerParameters, which spawns it as a subprocess, an attacker who signs up or uses stolen credentials gains full host takeover. A step-by-step and video proof-of-concept is public (publicly available exploit code exists), though there is no confirmed active exploitation.
Out-of-bounds write in the Zephyr RTOS ADIN2111/ADIN1110 Ethernet driver (versions 3.7.0-4.4.x) allows an adjacent-network attacker to corrupt up to ~14.8 KB of kernel memory by sending crafted Ethernet frames over a 10BASE-T1S or 10BASE-T1L single-pair segment. The driver's OPEN Alliance SPI frame-reassembly function copies attacker-controlled 64-byte chunks into a fixed 1524-byte static buffer without validating the write cursor against the buffer boundary, enabling writes into adjacent driver and kernel memory. SSVC indicates a proof-of-concept exists; however, EPSS is low (0.23%, 14th percentile) and CISA KEV does not list active exploitation, suggesting the flaw is known but not yet widely weaponized.
Remote code execution in PyTorch Lightning through 2.6.5 allows an attacker who can get a victim to load a malicious checkpoint file to execute arbitrary code. The flaw lives in the _load_state routine, which imports and calls the module path named in a checkpoint's _instantiator hyperparameter, letting a crafted .ckpt bypass torch's weights_only=True safeguard when LightningModule.load_from_checkpoint is invoked. Reported by VulnCheck, it is fixed in commit d710d68 (PR #21832) via an instantiator allowlist, and publicly available exploit code exists though it is not listed in CISA KEV.
Authenticated remote code execution in MantisBT versions 1.3.0 through 2.28.3 allows an administrator to run arbitrary PHP as the web server user (www-data) via the 'Manage Configuration' page (adm_config_set.php). When a config value is stored with a non-string type, the ConfigParser/Tokenizer path calls eval() on attacker-controlled code guarded only by a 'return;' prefix; because PHP hoists class and function declarations at compile time regardless of the return, an attacker can plant a class that later hijacks the autoloader. There is no public exploit identified at time of analysis and it is not listed in CISA KEV; the flaw was reported privately by watchTowr and fixed in 2.28.4.
Remote code execution in Metabase (open-source business intelligence platform) versions 1.54.0 through the fixed releases lets a user holding database-connection management rights run arbitrary code on the Metabase server. By pointing a Snowflake connection at an attacker-controlled server, a flaw in the bundled Snowflake JDBC driver writes attacker files anywhere on the host, overwriting one of Metabase's own driver files that later executes inside the Metabase process. There is no public exploit identified at time of analysis, but the vendor rates it CVSS 10.0 and a vendor security advisory (GHSA-r6x2-rchx-q9g9) confirms both the flaw and fixed versions.
Command injection in Tabby (formerly Terminus) terminal emulator before 1.0.234 lets an attacker achieve code execution by tricking a victim into dragging and dropping a crafted file whose path contains shell command-substitution metacharacters ($(...) or backticks). Because the drop handler in tabby-electron/src/pathDrop.ts inserted the raw path into the active shell without neutralizing these characters, the payload runs when the victim presses Enter. This is an incomplete-fix follow-up to CVE-2026-45038, which only addressed control characters; no public exploit is identified at time of analysis and it is not listed in CISA KEV.
Heap buffer overflow in NGINX Plus and NGINX Open Source lets unauthenticated remote attackers crash worker processes (DoS) and, on hosts with ASLR disabled or bypassed, potentially execute arbitrary code by sending crafted HTTP requests. The flaw is a data-plane-only issue triggered when a regex-based map directive references the map's regex capture variables before the map output variable in a string expression, or via a non-cacheable variable under certain conditions. F5 has released a patched version; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Server-side request forgery leading to code execution in Cursor's browser-enabled Cloud Agent lets attacker-controlled web content reach an unauthenticated local agent endpoint from inside the sandbox, compromising the session. Affecting Cursor Cloud Agent sessions prior to the 03/31/2026 fix, an attacker who gets a browsing agent to load malicious content can run code in the sandbox and read repository contents, environment variables, credentials, and GitHub App access tokens scoped to that session. No public exploit identified at time of analysis, and it is not listed in CISA KEV; the flaw is a missing-authentication (CWE-306) authentication-bypass class issue with high confidentiality, integrity, and availability impact to the affected session.
Arbitrary code execution in PraisonAI (praisonaiagents) before 1.6.78 lets an attacker who can drop a Python file into a plugin directory run their own code. The plugin manager auto-discovers and executes any .py file found in project-level or user-home .praisonai/plugins/ directories at initialization, with no code signing, integrity checks, or sandboxing. VulnCheck reported it under CWE-94; there is no public exploit identified at time of analysis and it is not listed in CISA KEV.
Privilege-scoped remote code execution in SolarWinds Serv-U managed file transfer server stems from an insecure direct object reference that lets an authenticated domain administrator reference objects outside their intended authorization boundary to run code on the host. The flaw carries a 9.1 CVSS due to network reach, scope change, and full confidentiality/integrity/availability impact, though the practical bar is high because domain administrator access is a prerequisite, and impact is reduced on Windows deployments. No public exploit is identified at time of analysis and it is not listed in CISA KEV.
Privilege-to-root remote code execution in SolarWinds Serv-U managed file transfer server stems from an insecure direct object reference (CWE-639) that an authenticated domain administrator with read/write access to the home directory can abuse to execute code as root on the underlying host. The CVSS 3.1 score of 9.1 reflects a scope change from application-admin context to full OS compromise, with the impact explicitly noted as lower on Windows deployments. There is no public exploit identified at time of analysis and the issue is not listed in CISA KEV, so this is a high-severity but authentication-gated flaw rather than a mass-exploitable unauthenticated bug.
Remote code execution in SolarWinds Serv-U managed file transfer/FTP server allows an attacker to run arbitrary code remotely, executing as root on Unix/Linux deployments while impact is reduced on Windows. The flaw stems from an access-control weakness (CWE-284) and is corroborated by a vendor 'Authentication Bypass' tag, carrying a CVSS 9.1 (scope-changed) rating. No public exploit identified at time of analysis and it is not listed in CISA KEV, so exploitation is currently theoretical rather than confirmed.
Privilege escalation to remote code execution as root in SolarWinds Serv-U stems from an insecure direct object reference (IDOR) that lets an authenticated group administrator manipulate object references beyond their authority. The flaw carries a CVSS 9.1 rating because a scope change allows a lower-tier admin to fully compromise the underlying host on Linux/Unix, though impact is reduced on Windows deployments. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, but the high privilege prerequisite (group admin) is realistic in multi-tenant MFT environments.
Local privilege escalation in snap-confine - the SUID/set-capabilities helper snapd uses to build the confinement sandbox for snap applications - lets an unprivileged local user on Ubuntu 22.04, 24.04, and 26.04 LTS gain full root. The flaw only manifests where snap-confine is deployed with file capabilities (set-capabilities) instead of the traditional set-uid-root binary, where incorrect handling of ambient capabilities lets an attacker escape intended privilege boundaries and run arbitrary code as root. Reported by Canonical; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Root command injection in dracut's NetworkManager-based initrd network module allows an adjacent-network attacker to execute arbitrary code inside the initramfs during early boot. By answering a booting system's DHCP request with malicious root-path, next-server, or bootfile-name options, the attacker's payload is written unescaped into a temporary shell script and executed as root before the real OS starts. No public exploit is identified at time of analysis, and the flaw is not on CISA KEV, but a Red Hat advisory (RHSA-2026:26534) and upstream fix commit confirm the issue is real and patched.
Memory-corruption in Mozilla Firefox ESR 115.37 and 140.12 could allow attackers to run arbitrary code by triggering memory-safety bugs (CWE-119) that Mozilla observed showing evidence of memory corruption. Reported by Mozilla and tracked under advisories MFSA2026-69/70, the flaw is remotely reachable and carries a CVSS 9.8, though there is no public exploit identified at time of analysis and no active-exploitation (KEV) listing. Mozilla explicitly frames exploitation as presumptive ('with enough effort'), indicating no confirmed working exploit.
Memory-safety defects in Mozilla Firefox 152 and the Firefox ESR 115.37 and 140.12 release trains can corrupt process memory, and Mozilla states some showed evidence of memory corruption that could be leveraged to run arbitrary code. The flaws were resolved in Firefox 153, Firefox ESR 115.38, and Firefox ESR 140.13. There is no public exploit identified at time of analysis, and the EPSS probability is low (0.14%, 4th percentile), consistent with a batch of internally-found bugs rather than an actively exploited zero-day.
Memory corruption in Mozilla Firefox 152 and Firefox ESR 140.12 could allow attackers to run arbitrary code within the browser, per Mozilla advisories MFSA2026-68 and MFSA2026-70. The flaw is a batch of internally-discovered memory safety bugs (CWE-119) affecting the Firefox rendering/JavaScript engine, at least some of which showed evidence of exploitable memory corruption. No public exploit identified at time of analysis; the issues were reported by Mozilla itself and fixed in Firefox 153 and Firefox ESR 140.13.
Memory corruption in Mozilla Firefox 152 lets a remote attacker who lures a victim to a malicious web page trigger memory-safety violations that Mozilla assesses could be leveraged for arbitrary code execution. The flaw is one of a batch of internally-discovered memory safety bugs (CWE-119) rolled up under MFSA2026-68 and carries a CVSS 9.8 rating; there is no public exploit identified at time of analysis, and it is fixed in Firefox 153.
Remote code execution in Grav CMS 2.0.4 (fixed in 2.0.7) lets an authenticated user holding the admin.pages or api.pages.write permission plant a malicious Class::method callable directive in page frontmatter, which Blueprint::dynamicData() feeds unfiltered into call_user_func_array(). The injected command then executes as the web-server user for any subsequent visitor to the page, including unauthenticated ones. Reported by VulnCheck (CWE-94, CVSS 4.0 base 9.3); no public exploit identified at time of analysis and it is not listed in CISA KEV.
Stack-based buffer overflow in libssh's SFTP server allows a connected client to crash the server or potentially execute code by placing files with overlong names into a directory the server subsequently lists. The flaw lives in how libssh builds the SFTP 'longname' field, concatenating attacker-influenced filenames into a fixed-size stack buffer without adequate bounds checking. There is no public exploit identified at time of analysis and EPSS is very low (0.10%), but the memory-corruption nature (CWE-121) and libssh's broad deployment make it worth prioritized patching.
Arbitrary file upload in the MapSVG WordPress plugin (all versions through 8.14.0) lets authenticated Administrator-level users bypass file-type validation in the SVGFile constructor and upload arbitrary files, potentially achieving remote code execution on the host. The flaw stems from an incorrect conditional that skips validation entirely rather than enforcing it. Reported by Wordfence with a CVSS 3.1 base score of 7.2; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Arbitrary code execution in DayuanJiang next-ai-draw-io 0.4.13 arises from a path traversal flaw in the bundled MCP (Model Context Protocol) server, allowing an attacker who can influence a request to the mcp-server component to write or access files outside the intended directory and ultimately run code on the host. The affected product is a niche open-source AI-assisted draw.io diagramming application. No public exploit has been confirmed, though a third-party technical writeup exists; EPSS is low (0.43%, 35th percentile) and the flaw is not in CISA KEV.
Cross-site scripting in DayuanJiang next-ai-draw-io 0.4.13 lets a remote unauthenticated attacker inject arbitrary script through the mcp parameter, executing in the victim's browser when they open a crafted link. The 'arbitrary code' claim and RCE tag in the source data most plausibly refer to arbitrary JavaScript in the browser (CWE-79 reflected XSS) rather than server-side code execution. EPSS is low (0.29%, 21st percentile), there is no CISA KEV listing, and no public exploit is confirmed, though third-party GitHub writeups referencing this CVE exist.
Remote arbitrary code execution in Bytebot (bytebot-ai self-hosted AI desktop agent, commit 3d37894 dated 2025-09-11) allows attackers to run OS commands by supplying a crafted path to the agent's `computer_write_file` action. The flaw is classed as OS command injection (CWE-78) and carries a CVSS 9.8, but real-world exploitation probability is currently low (EPSS 0.17%, 7th percentile) and it is not on CISA KEV. A public gist referenced by the CVE appears to document the issue and may contain proof-of-concept details, though POC status is not explicitly confirmed in the source data.
Privilege escalation to remote code execution in dotCMS 21.02 through 26.06.22-03 lets a low-privileged authenticated backend user abuse improper authorization checks in the ToolGroupResource (REST) and RoleAjax (DWR) endpoints to self-assign the administrative layout and self-grant the CMS Administrator role, then upload a crafted OSGi bundle whose BundleActivator runs arbitrary shell commands on the host. The endpoints failed to verify the caller was an admin before honoring layout- and role-assignment requests, so any backend account could bootstrap itself to full administrator. There is no public exploit identified at time of analysis, but an upstream fix (PR #36344) is available and the vendor-supplied CVSS 4.0 score is 9.4 (Critical).
Arbitrary code execution in Adobe DNG SDK version 1.7.1 (build 2536) and earlier arises from a stack-based buffer overflow triggered when parsing a maliciously crafted image file, allowing an attacker to run code in the context of the current user. Applications that embed the DNG SDK to decode Digital Negative (DNG) raw camera files are affected, and successful exploitation requires the victim to open the malicious file. No public exploit identified at time of analysis, and the flaw is not listed in CISA KEV.
Authenticated PHP code execution in ThemeXpert's Quix Page Builder Pro extension for Joomla lets a privileged builder user (holding core.create or core.edit permissions) embed raw PHP tags in element content that are subsequently executed when the server's view-cache layer include()s the cached template. Exploitation requires caching to be enabled, which is the default, and yields full remote code execution on the underlying web server. No public exploit identified at time of analysis, though the reporter is Joomla itself and the CVSS 4.0 base score is 8.9 (High).
Remote code execution in Chamilo LMS 1.11.40 and earlier allows any user enrolled in a course to run arbitrary PHP through the main/inc/ajax/lang.ajax.php endpoint, which is guarded only by api_protect_course_script(true) and is therefore reachable by students, teachers, and DRH (HR) roles rather than administrators. The root cause is an improper-authorization defect (CWE-285) that exposes a dangerous language-editing endpoint to low-privilege course members. No public exploit code is identified at time of analysis and it is not listed in CISA KEV, but a vendor security advisory (GHSA-4hwq-pv7c-3928) has been published.
Stack-based buffer overflow in rsyslog's optional mmpstrucdata module allows a remote, unauthenticated attacker to crash the rsyslog process and interrupt log collection, with theoretical code execution on weakly hardened builds. Affected versions span rsyslog 7.5.4 through 8.2605.x; the fix is included in 8.2606.0. No public exploit code or active exploitation has been confirmed, but the crash was independently reproduced over both TCP and UDP on Debian 12 with rsyslog 8.2302.0, and the corrective PR (#6991) was merged publicly before coordinated disclosure completed.
Remote code execution in Piwigo 16.3.0 and earlier lets an unauthenticated attacker inject arbitrary PHP through the web installer, which writes unsanitized POST parameters (prefix, dbpasswd, dbhost, dbname, dbuser) into local/config/database.inc.php where they execute on every page load. The bug is a PHP 8 regression: the addslashes() guard is gated on get_magic_quotes_gpc(), removed in PHP 8.0, so escaping is silently skipped and raw input reaches PHP source. No public exploit identified at time of analysis, but the flaw is trivially weaponizable when the installer is reachable, yielding full server compromise.
Arbitrary code execution in ProFTPD's mod_sftp module (versions before 1.3.9c and 1.3.10rc3) allows authenticated low-privilege SFTP users to run code in the server process by abusing SFTP packet reassembly in fxp.c. The flaw stems from an incorrectly conditioned reallocation of the 16 KB reassembly buffer that corrupts memory-pool freelist metadata, ultimately hijacking pr_fsio_stat() to reach system(). Reported by VulnCheck with a vendor patch available; no public exploit identified at time of analysis, though the advisory describes a complete exploitation primitive.
Remote code execution in Apache Syncope's connector subsystem allows an administrator holding sufficient entitlements to run arbitrary Groovy scripts through scripted REST and SQL connectors, escaping the intended sandbox and executing code on the host. It affects Syncope 3.0.0-M0 through 3.0.16, 4.0.0-M0 through 4.0.6, and 4.1.0-M0 through 4.1.1, and is rooted in improper isolation (CWE-653) of the scripting engine. Although the published CVSS is 9.8, the vendor description states an authenticated privileged administrator is required, and there is no public exploit identified at time of analysis.
Local privilege escalation to root in the QEMU Guest Agent (qga) affects Red Hat Enterprise Linux 6 through 10, RHEL for NVIDIA 26, and OpenShift Container Platform 4, where a local unprivileged guest user abuses the guest-ssh-add-authorized-keys command handler through symlink manipulation to seize ownership of arbitrary root-owned files. Two attack variants exist - a deterministic directory-symlink bypass and a TOCTOU file-symlink race - both requiring an external management layer such as libvirt to invoke the code path. No public exploit is identified at time of analysis and it is not listed in CISA KEV, but the CVSS 7.3 rating and clean path to full root access make it a meaningful local escalation risk in multi-tenant virtualization estates.
Path traversal in Menyoo (MenyooSP), a popular GTA V single-player trainer/mod, before commit 729aa48 lets a local user supply crafted file names (containing traversal sequences) to the Spooner, VehicleSpawner, WeaponOptions, and PedComponentChanger save/create-folder/rename functions, writing or renaming files outside the intended data directories. Because the resulting XML/config files can be planted in attacker-chosen locations, this escalates to arbitrary code execution in the victim's context. There is no public exploit identified at time of analysis and it is not in CISA KEV; the fix (added IsSafePath() input validation) is publicly visible in PR #550.
Arbitrary code execution in SJCAM SJ4000-Air action cameras (firmware V1.4C and earlier), and unbranded whitelabel devices built on the same v1.4C firmware, is possible via a maliciously crafted FEX file - the Allwinner firmware/partition configuration format used by these AllwinnerTech-based devices. The flaw is classified as code injection (CWE-94) with a supplied CVSS of 9.8; no public exploit or CISA KEV listing exists at time of analysis, though a detailed reverse-engineering writeup of the device firmware is publicly referenced. The AV:N network rating in the supplied vector is inconsistent with the description's file-processing mechanism and warrants verification.
Poisoned-pipeline (pwn request) code execution in the Meshtastic firmware repository's GitHub Actions CI allows any external user to run arbitrary code inside the project's trusted CI context prior to firmware 2.7.21.1370b23. Because main_matrix.yml (and related workflows) triggered on pull_request_target while the check, build, and build-debian-src jobs checked out and executed code from the attacker's fork with no approval gate, a pull request from an author_association of NONE automatically ran attacker-controlled files with access to repository secrets and an elevated GITHUB_TOKEN. This is CWE-94 code injection scoring CVSS 10.0; no public exploit identified at time of analysis, though a linked Google Drive writeup may contain proof-of-concept detail.
Arbitrary code execution in keras-team/keras 3.15.0 arises because the public keras.layers.TorchModuleWrapper.from_config method calls torch.load(..., weights_only=False) without any explicit unsafe opt-in, so a maliciously crafted Keras layer configuration deserialized outside a SafeModeScope(True) context triggers execution of attacker-controlled PyTorch pickle payloads (CWE-502). Any application or pipeline that loads untrusted or third-party Keras model/layer configs is affected; CVSS is 7.8 (High) with a local vector requiring user interaction. There is no public exploit identified at time of analysis and the flaw is not listed in CISA KEV, but the vulnerability class is well understood and reliably weaponizable via standard pickle reduce gadgets.
Session hijacking and keystroke injection in the urwid Python library's web display backend (urwid/display/web.py) let attackers take over active web terminal sessions and reach OS-level code execution with the session owner's privileges. The urwid_id session token is generated with Python's non-cryptographic Mersenne Twister PRNG, so a remote attacker who observes roughly 334 session IDs (e.g. via the X-Urwid-ID response header) can reconstruct PRNG state and predict every past and future token, while any local user can read the same token from world-listable FIFO filenames in /tmp. No public exploit has been identified at time of analysis; the issue was reported by VulnCheck and carries a 9.2 (CVSS 4.0) severity.
Arbitrary Python code execution in scikit-hep uproot (uproot5) arises because the library dynamically generates and compiles reader classes from a ROOT file's TStreamerInfo records, interpolating file-controlled streamer metadata (such as streamer element names) into generated Python source without repr()/!r safe quoting. A crafted ROOT file can inject expression-breaking content that is evaluated in the process opening the file, giving code execution to whoever processes attacker-supplied ROOT data. Reported by VulnCheck; no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Remote code execution in VMware Avi Load Balancer exposes organizations running versions from 22.1.1 through 32.1.1 to arbitrary code injection by low-privileged authenticated users with network access. The flaw is classified as CWE-94 code injection, meaning attacker-supplied input reaches an execution context without adequate sanitization. Broadcom has released patches across all affected branches; no public exploit code has been identified at time of analysis, and the EPSS score of 0.68% (48th percentile) reflects modest current exploitation probability despite the high CVSS 8.8 rating.
Remote code execution in VMware Avi Load Balancer exposes the Avi Control Plane to arbitrary code execution by any low-privileged, network-connected user, stemming from a code injection flaw (CWE-94) in the controller. Multiple major version branches from 22.1.1 through 32.1.1 are confirmed affected, with vendor-released patches available across all branches. No public exploit code or CISA KEV listing has been identified at time of analysis, though the CVSS 8.8 rating and full high-impact triad (C:H/I:H/A:H) make this a high-priority patch target for any deployment where the control plane is network-reachable.
Remote code execution and information disclosure in IBM Db2 Genius Hub (1.1, 1.1.1, 1.1.2) and IBM Agentics 1.0 stem from the use of inherently dangerous functions invoked without sufficient input restrictions, allowing remote attackers to run arbitrary code or read sensitive data. IBM's PSIRT reported the flaw with a 9.8 critical CVSS reflecting network-reachable, unauthenticated exploitation with full confidentiality, integrity and availability impact. There is no public exploit identified at time of analysis, and the low EPSS score (0.33%, 25th percentile) suggests exploitation is not yet widespread.
Heap-based buffer overflow in IBM Storage Protect Client (versions 8.1.0.0-8.1.27.1 and 8.2.0.0-8.2.1.0) lets a remote attacker corrupt heap memory through improper bounds checking, potentially achieving arbitrary code execution or crashing the affected process. The flaw carries a CVSS 3.1 base score of 9.8, but no public exploit identified at time of analysis and the EPSS probability is low (0.46%, 37th percentile). Note that the affected component is the backup Client, while the description references crashing 'the server,' introducing ambiguity about the true exploitation path.
Unauthenticated remote code execution in IBM Langflow OSS versions 1.0.0 through 1.10.1 lets anonymous attackers run arbitrary code by abusing the public flow build endpoint (/api/v1/build_public_tmp/{flow_id}/flow). The CVSS 3.1 base score is 9.8 (network, low complexity, no privileges or user interaction). No public exploit is identified at time of analysis, and the EPSS probability is modest at 0.44% (36th percentile), indicating exploitation is technically trivial but not yet observed at scale.
Extension-denylist bypass in the Flask-Reuploaded (flask_uploads) Python library allows an attacker who controls the name override passed to UploadSet.save() to store dangerous script files by varying letter case (e.g. shell.PHP, evil.pHp), defeating the very script-blocking denylist the library documents. This is an incomplete-fix variant of CVE-2026-27641: the v1.5.0 fix re-validated overridden names through the case-preserving extension() helper instead of the case-normalizing lowercase_ext() used by the default path, so uppercase extensions slip past a lowercase-tokened denylist. Publicly available exploit code exists (a self-contained PoC verified against Flask-Reuploaded==1.5.0), though there is no evidence of active exploitation and no EPSS score was provided.
Arbitrary code execution in Adobe Acrobat Reader arises from a heap-based buffer overflow (CWE-122) triggered when a victim opens a maliciously crafted file, allowing an attacker to run code in the context of the current user. All affected versions covered by Adobe advisory APSB26-63 are impacted, and successful exploitation grants full high impact to confidentiality, integrity, and availability. There is no public exploit identified at time of analysis, and the issue is not listed in CISA KEV, so risk is currently latent rather than active.
Remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows an authenticated remote attacker to run arbitrary code because validation is not fully enforced on Model Context Protocol (MCP) server configuration files, permitting injection of attacker-controlled commands or server definitions. IBM has published a fix advisory (node/7278932). With CVSS 8.8 (PR:L), a low-privileged user of the Langflow instance can escalate to full code execution on the host; there is no public exploit identified at time of analysis and it is not listed in CISA KEV.
Remote code execution in WordPress 6.9.x (before 6.9.5) and 7.0.x (before 7.0.2) arises from a REST API batch endpoint route confusion flaw that, when chained with the author__not_in WP_Query SQL injection (CVE-2026-60137), lets attackers inject SQL and ultimately execute arbitrary code on the underlying host. Publicly available exploit code exists (a ProjectDiscovery Nuclei template and WPScan advisory), and the CISA SSVC framework rates the flaw as automatable with total technical impact, though it is not currently listed in CISA KEV. The vulnerability was reported by WPScan and patched in the WordPress 7.0.2 and 6.9.5 releases.
Remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows authenticated users to override arbitrary component parameters at runtime through the API, bypassing the parameter filtering logic in the apply_tweaks() function (CWE-94 code injection). Because a low-privileged authenticated account is sufficient and the impact is full compromise of confidentiality, integrity, and availability (CVSS 8.8), this maps to server-side code execution on the Langflow host. There is no public exploit identified at time of analysis, and the vulnerability is not listed in CISA KEV.
Remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows attackers to run arbitrary code by planting a malicious pickle payload that the AsyncDiskCache deserializes with unauthenticated pickle.loads(). Any actor able to influence cached data - via file system access, crafted workflow inputs, custom components, or API manipulation - gains full compromise at the privilege of the Langflow server process. Rated CVSS 9.9 and reported by IBM with a vendor patch available; no public exploit identified at time of analysis.
Authenticated remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 lets any logged-in user run arbitrary Python — and therefore arbitrary OS commands — with the privileges of the Langflow server process. The flaw is a design failure in the POST /api/v1/validate/code endpoint, which passes user-supplied code straight into Python's exec() with no sandbox, validation, or privilege drop. Rated CVSS 9.9 by IBM; no public exploit identified at time of analysis, though the trivial exec()-based trigger makes exploitation straightforward once authenticated.
Unauthenticated remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 stems from broken webhook authentication that skips API key validation whenever WEBHOOK_AUTH_ENABLE is False - which is the shipped default. Any remote attacker who learns a flow's UUID can invoke that flow's webhook endpoint as though they were its owner, running arbitrary flow logic and reaching RCE. No public exploit identified at time of analysis, but the CVSS 9.8 rating and default-vulnerable configuration make this a high-priority patch.
Privilege escalation to remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows an authenticated user to elevate their account to superuser by directly manipulating the underlying database, then execute arbitrary system commands and achieve full compromise of the host running the Langflow service. The flaw (CWE-94, code injection) carries a critical CVSS 9.9 with a scope change, meaning code runs with the Langflow service account's permissions beyond the vulnerable component. No public exploit identified at time of analysis, and it is not listed in CISA KEV, but a vendor patch is available.
Arbitrary Python code execution in IBM Langflow OSS 1.0.0 through 1.10.0 (Langflow versions up to 1.9.2) allows authenticated users with flow-creation privileges to bypass the allow_custom_components=false control via the Policies component's ToolGuard integration. Because validation only inspects the main component code and ignores dynamic CodeInput fields, attackers persist malicious Python in a flow that runs server-side when a guarded tool executes. No public exploit has been identified at time of analysis, but the flaw is IBM-confirmed with a vendor patch and carries a critical 9.9 CVSS due to cross-tenant escalation through the agentic MCP update_flow_component_field tool.
Unauthenticated remote code execution in IBM Langflow OSS versions 1.0.0 through 1.10.0 lets any network-reachable attacker achieve full RCE on default deployments by chaining two endpoints: /api/v1/auto_login, which mints SUPERUSER tokens to any caller, and /api/v1/validate/code, which passes user-supplied code to Python's exec(). CVSS is 9.8 (AV:N/AC:L/PR:N/UI:N) and a vendor patch is available; there is no public exploit identified at time of analysis, and the flaw is not listed in CISA KEV.
Code injection leading to remote code execution affects IBM Db2 versions 11.5.0 through 11.5.9 and 12.1.0 through 12.1.4 when an attacker can influence the JDBC connection URL passed to the driver. An attacker who controls the JDBC URL (typically via an application that builds connection strings from user input) can inject malicious properties or references that cause arbitrary code to execute in the context of the connecting application/driver host. IBM has released a fix; there is no public exploit identified at time of analysis and the issue is not listed in CISA KEV.
Arbitrary code execution in ForgeCode (tailcallhq/forgecode), an AI pair-programming CLI, occurs because the tool auto-loads and honors a repository's project-local .mcp.json at startup without asking the user. A malicious repository can define mcpServers entries with arbitrary command/args (e.g. bash -c), so simply running the forge CLI inside a cloned untrusted repo spawns attacker-chosen commands as the invoking developer. Reported by VulnCheck with a vendor patch available; no public exploit tool is identified at time of analysis, though the CVE description itself embeds a working payload example.
Unauthenticated SQL injection in Sangoma Switchvox SMB Edition 8.3 (build 104997) lets remote attackers execute arbitrary SQL against the backend PostgreSQL database via the /pa endpoint, which concatenates the attacker-supplied PhoneIP value from a <PolycomIPPhone> XML body into queries without parameterization. Because SQL access on this PostgreSQL deployment can be escalated to command execution, a single crafted, unauthenticated request yields full database compromise and remote code execution on the VoIP appliance. There is no public exploit identified at time of analysis, though vendor SRA published a technical writeup, and the CVSS 4.0 base score is 9.3 (Critical).
Arbitrary file write in GitHub Enterprise Server (all releases before 3.22) lets an attacker who already has code execution inside the Dependabot updater container escape the intended dependency-file directory via path traversal and symlink manipulation, planting attacker-controlled files anywhere in a repository - including GitHub Actions workflows under .github/workflows/. Because path validation checked the declared rather than the effective (symlink-resolved) path, an injected workflow can run with the repository's Actions secrets when the repo uses pull_request_target or auto-merge. There is no public exploit identified at time of analysis and it is not listed in CISA KEV; CVSS 4.0 base score is 8.6 (High).
Arbitrary code execution in Cursor for Windows 3.2.16 lets remote attackers run code on a developer's machine by planting a malicious git.exe in a repository's root directory. Because Cursor resolves and executes the workspace-resident git.exe at IDE startup and on a recurring timer, simply cloning and opening a crafted repository triggers execution under the current user's privileges with no explicit user action beyond opening the project. Publicly available exploit code exists and the flaw was reported by VulnCheck; there is no CISA KEV listing or active-exploitation confirmation in the data.
Remote code execution is possible in the ProfilePress WordPress plugin (Paid Membership, Ecommerce, User Registration) in all versions through 4.16.18, where the DigitalProducts UploadHandler unconditionally registers an upload_mimes filter that whitelists executable extensions (.exe, .apk, .msi) across every WordPress upload context site-wide. Authenticated attackers with author-level access or above (per CVSS PR:L) can leverage this expanded MIME allowlist to upload executable files, leading to code execution. There is no public exploit identified at time of analysis, and this CVE is not listed in CISA KEV.
Remote code execution in the Generic OEM UZ801_v2.1 4G LTE Router firmware V3.4.3 allows an unauthenticated remote attacker to run arbitrary commands through the exposed sbin/adbd (Android Debug Bridge daemon) component. The CVSS 3.1 base score is 9.1 (AV:N/AC:L/PR:N/UI:N) with high confidentiality and integrity impact, reflecting full command execution on the device. No CISA KEV listing is present and EPSS is low (0.35%, 27th percentile), but a researcher GitHub repository is referenced, indicating publicly available exploit code likely exists.
Hardcoded default administrator password in Pheditor (all versions prior to 2.0.6) lets remote unauthenticated attackers log in as admin using the string 'admin' — the SHA-512 hash of which is baked into pheditor.php line 11 — and then abuse the built-in terminal and file-upload features for immediate remote code execution. Any deployment left on defaults is trivially compromised because there is no forced password change, expiry, lockout, or setup wizard. A working PoC is published in the GitHub Security Advisory, though there is no public exploit identified in active in-the-wild campaigns and no EPSS/KEV data is provided.
Build-time remote code execution in Nuclio (serverless/FaaS platform) versions <= 1.15.27 lets remote attackers run arbitrary OS commands as root inside the function-builder container. The Java runtime renders user-supplied runtimeAttributes.repositories values into a Groovy build.gradle file via Go's text/template using the non-escaping {{ . }} action, so an attacker can inject a closing brace to break out of the repositories {} block and append Groovy that Gradle executes during its configuration phase. Because the Dashboard API defaults to NOP (no) authentication, exploitation needs no credentials; a dynamically-verified proof-of-concept is published in the vendor advisory, though there is no public exploit identified as being used in active attacks.
Remote kernel heap corruption in the illumos SCTP stack lets an unauthenticated attacker send a crafted INIT ACK packet with malformed address parameters to trigger an out-of-bounds access during packet classification, potentially leading to remote code execution or a kernel panic. The flaw affects illumos-gate and downstream distributions (OmniOS, SmartOS/Triton) built before commit 53a3efde, and has existed since 2010 (commit a5407c02). There is no public exploit identified at time of analysis and the CVSS 4.0 exploit-maturity metric is Unreported (E:U), but the attack requires no authentication or user interaction.
Arbitrary file disclosure in Envoy Gateway's controller allows a low-privileged tenant who can create an EnvoyExtensionPolicy to read sensitive files from the gateway controller pod. A path-normalization gap in the security.lua critical-path check (CWE-20) lets attacker-supplied Lua bypass the read restriction using double-slash paths (e.g. //etc/passwd), exposing Kubernetes service-account tokens, TLS certificates, and process environment. No public exploit identified at time of analysis, but the vendor advisory documents the exact bypass and CVSS is rated 9.1.
Code injection in remorses/genql (the @genql/cli TypeScript GraphQL client generator) before 6.3.4 lets an actor who controls the GraphQL schema fed to genql smuggle arbitrary TypeScript/JavaScript into the generated schema.ts file, which then executes when a downstream consumer bundles and imports the client. The flaw stems from GraphQL type/field descriptions containing a `*/` sequence that breaks out of the generated JSDoc comment block. There is no public exploit identified at time of analysis and it is not in CISA KEV; the issue was responsibly disclosed (issue #197) and reported by CISA (cisa-cg), with a vendor patch in 6.3.4.
Code injection in Frogman before 1.6.2 lets an authenticated PERM_WRITE caller of the MCP/HTTP dialplan API inject arbitrary Asterisk directives - including System() and Set(SHELL(...)) - into extensions_custom.conf, yielding remote command execution on the PBX host. The flaw stems from fm_dialplan_apply sanitizing only the context name while passing template parameters such as greeting, dest, url, extension, code, and file into generated dialplan text unchecked. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV, but the CVSS is 9.9 and the fix (v1.6.2) plus root-cause commit are public.
Local privilege escalation to arbitrary code execution affects Lenovo App Store, a software-distribution client shipped exclusively on Lenovo devices in the Chinese market. A local authenticated user who can interact with the application can leverage its excessive privileges (CWE-250) to run arbitrary code at an elevated privilege level, with high confidentiality, integrity, and availability impact on the host. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, so no active exploitation is confirmed.
Local code execution in Lenovo App Store (Chinese-market distribution only) lets an authenticated local user abuse a path traversal weakness to write or load files outside intended directories and run arbitrary code. The flaw was self-reported by Lenovo and carries a CVSS 4.0 base score of 7.0; there is no public exploit identified at time of analysis and it is not listed in CISA KEV. Impact is high to confidentiality, integrity, and availability of the local system, but exploitation requires existing low-privilege local access plus user interaction.
Local code execution in Lenovo Legion Zone and the Lenovo App Store (China-market Windows builds) allows an authenticated local user to run arbitrary code when either application is installed on a non-system partition. The flaw stems from insecure directory/file permissions (CWE-277) that a low-privileged user can abuse to plant or modify executable content later run with higher privilege. No public exploit is identified at time of analysis and it is not listed in CISA KEV; impact is confined to systems where these Chinese-market Lenovo utilities are deployed to a non-default drive.
Arbitrary JavaScript execution in Microsoft's @prompty/core TypeScript loader (2.0.0-alpha.1 through 2.0.0-beta.2) allows an attacker who supplies a crafted .prompty file to run code during prompt loading. The loader passed .prompty markdown to gray-matter without disabling its executable 'js'/'javascript' frontmatter engines, so a '---js' frontmatter block is evaluated as Node.js at parse time. Exploitation requires the victim's application to load the malicious file (UI:P); no public exploit is identified at time of analysis and it is not on CISA KEV, but the fix commit ships a working proof-of-concept test.
Unauthenticated remote code execution in SGLang (versions 0 through 0.5.14) arises when the expert-parallel backup subsystem binds a ZeroMQ PULL socket to a routable interface without authentication or safe deserialization, letting a network attacker send a crafted pickle payload that executes arbitrary code in the serving process. It affects deployments where the elastic expert-parallel backup feature is enabled and the socket is reachable. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Command injection in Microsoft Kiota before 1.32.5 lets a malicious or compromised OpenAPI description dictate the install command that Kiota presents to developers. When a developer runs `kiota info` (or the VS Code extension's `kiota info --json` dependency-install flow) against an attacker-controlled spec, Kiota reads the `x-ms-kiota-info.languagesInformation.<language>.dependencyInstallCommand` field plus attacker-chosen dependency name/version values and surfaces them as its trusted recommended install command, achieving arbitrary command execution when that command is run. No public exploit has been identified at time of analysis and it is not in CISA KEV, but the fix explicitly removes the untrusted extension field.
PHP code generation in Microsoft Kiota prior to 1.32.4 permits injection of arbitrary PHP code when the generator processes attacker-controlled OpenAPI specification fields, because the shared `SanitizeDoubleQuote()` helper in `Writers/StringExtensions.cs` fails to escape the `$` character before embedding schema-derived strings into PHP double-quoted literals. Developers or CI/CD pipelines that generate PHP API clients from untrusted or third-party specifications will produce PHP files containing live interpolation expressions (`${...}`, `$var`, `{$obj->prop}`) that execute attacker-chosen code at PHP runtime in production environments. No public exploit has been identified and SSVC rates exploitation as currently none, but technical impact is rated total; the realistic risk is tightly gated on whether teams consume external or attacker-influenced specs without generated-code review.
Code injection in Microsoft Kiota's Python client generator (versions prior to 1.32.0) allows an attacker who controls an OpenAPI description to embed arbitrary Python that executes when a developer generates and imports the resulting client. The flaw stems from unsanitized x-ms-enum.values[].description strings being emitted as inline comments; embedded newlines break out of the comment and run at module scope. No public exploit identified at time of analysis, and the issue is not in CISA KEV. The assigned CVSS 3.1 base score is 7.5, though its impact metrics (C:N/I:N/A:H) understate what is functionally full code execution.
Code injection in Microsoft Kiota Ruby code generator allows remote code execution through crafted OpenAPI specifications. Kiota versions prior to 1.32.0 fail to escape '#' characters in schema-derived strings embedded in Ruby double-quoted literals, enabling Ruby interpolation sequences such as '#{code}' to execute arbitrary commands when the generated client code runs. While no active exploitation is confirmed, the attack is automatable, EPSS indicates moderate probability (1.47%), and the vendor has released a fix.
Code-generation injection in Microsoft Kiota's C# XML documentation-comment writer allows an attacker who controls or compromises an OpenAPI specification to inject arbitrary C# statements into generated client source files. Affected packages Microsoft.OpenApi.Kiota and Microsoft.OpenApi.Kiota.Builder in versions prior to 1.29.1 and in the 1.30.0-1.32.2 range fail to strip C# line-terminator characters from description and externalDocs fields before emitting them as single-line /// doc comments, enabling payload breakout into compilable namespace-level code. No public exploit confirmed in the wild at time of analysis, though SSVC reports a POC exists and rates technical impact as total; exploitation culminates when a developer or CI pipeline builds the tainted generated output.
Sandbox escape in PipeWire's PulseAudio compatibility layer lets a low-privileged process inside a confined environment such as Flatpak load an attacker-controlled library and execute arbitrary code on the host user's session. The flaw (CWE-427) is present in PipeWire as shipped in Red Hat Enterprise Linux 7, 8, 9 and 10 and carries a scope-changing CVSS 3.1 score of 8.8. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Root-level remote code execution in Canonical's ubuntu-pro-client (formerly ubuntu-advantage-tools) allows an attacker who can spoof or tamper with the Ubuntu Pro contract server response to inject arbitrary APT source lines and install malicious root-owned packages. The client builds /etc/apt/sources.list.d entries from the server's directives.suites[] and directives.aptURL fields via Python str.format() without escaping or newline filtering, and passes additionalPackages[] positionally into a root-run apt-get install. Because the component is preinstalled on supported Ubuntu Server releases and auto-attaches on cloud Ubuntu Pro images the exposure is broad; there is no public exploit identified at time of analysis and it is not listed in CISA KEV.
Arbitrary file deletion in the Uncanny Automator WordPress plugin (all versions through 7.3.1.4) lets unauthenticated attackers delete any file on the server via untrusted PHP object deserialization in the fr_token function, and deleting a critical file such as wp-config.php can pivute WordPress into an installation/setup state that yields remote code execution. Exploitation is gated by a specific setup: a Forminator form must be connected to an Uncanny Automator recipe whose trigger is configured for 'Everyone', which is what exposes the deserialization sink to anonymous form submissions. A self-contained gadget chain via Action_Helpers_Email::__destruct() ships inside the plugin, so no external gadget library is needed; reported by Wordfence with no public exploit identified at time of analysis and no CISA KEV listing.
Missing authentication in @andrea9293/mcp-documentation-server v1.13.0 exposes its document-management Web UI/API on all network interfaces (0.0.0.0:3080) instead of localhost, letting any adjacent-network client read, add, search, and delete documents in the MCP knowledge base without credentials. Rated CVSS 8.8 (CWE-306), the flaw is exploitable by unauthenticated attackers reachable over the same LAN, VM network, Docker bridge, or VPN. A detailed proof-of-concept exists in the reporter's advisory, though there is no public exploit identified as being used in active attacks and it is not listed in CISA KEV; the vendor fixed it in v1.13.1.
Sandbox escape with SYSTEM-level code execution in Sandboxie-Plus before 1.17.6 allows a process confined inside a sandbox to run arbitrary code in an unsandboxed host process. The privileged GuiServer service trusts attacker-supplied thread and function-pointer values from a GUI window-hook registration request and dispatches them via QueueUserAPC while running as SYSTEM, defeating the core isolation guarantee of the product. No public exploit identified at time of analysis and the flaw is not listed in CISA KEV, but the mechanism is a full sandbox-to-host escape.
Authenticated remote code execution in 9Router before 0.5.2 lets a logged-in attacker run arbitrary OS commands on the host by chaining a Host-header spoof (to reach routes that are supposed to be localhost-only) with unsanitized MCP plugin arguments that flow into child_process.spawn() via the /api/mcp//sse endpoint. Rated CVSS 8.8 (CWE-78, OS command injection), it is fixed in 0.5.2; no public exploit or CISA KEV listing exists at time of analysis, though a vendor advisory and fix commit are published.
Remote authenticated code execution in DataEase before 2.10.23 lets a low-privileged user abuse the Excel/datasource upload workflow to run arbitrary Java. By uploading a crafted payload.zip via /datasource/upload and defining an H2 datasource that references the archive through the zip: protocol, an attacker forces CalciteProvider.jdbcFetchResultField to execute a query that triggers precompiled Java aliases embedded in test.mv.db. The vendor CVSS 4.0 vector scores this 7.1 (confidentiality-only), but the underlying primitive is full arbitrary code execution; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Reflection-based remote code execution in DataEase before 2.10.23 lets an authenticated operator who can configure a Redshift datasource achieve arbitrary code execution on the server. The Redshift JDBC driver loads an attacker-controlled rsjdbc.ini from the system temp directory and honors a malicious socketFactory pointing at Spring's FileSystemXmlApplicationContext, turning a normal JDBC connection into a code-execution chain. No public exploit is identified at time of analysis and it is not in CISA KEV, but the vendor has confirmed and patched the flaw in 2.10.23.
Server-side arbitrary code execution in TDengine time-series database (versions prior to 3.4.1.15) allows a database user holding the 'create udf' privilege to upload a malicious shared library, register it as a user-defined function (e.g. named 'eval'), and execute attacker-controlled native C code on the server by invoking that function through ordinary SQL queries. This is a privileged-user code injection issue (CWE-94) with high confidentiality, integrity, and availability impact; no public exploit identified at time of analysis and it is not listed in CISA KEV. The flaw is fixed in version 3.4.1.15.
Stack-based buffer overflow in TDengine (open-source IoT time-series database) versions 3.4.1.6 and earlier allows an authenticated user to trigger a one-byte out-of-bounds write by submitting SQL containing crafted escape sequences (\%, \_, or \x), leading to denial of service and potentially remote code execution. The flaw sits in the SQL parser's trimString() routine, which validates only a single byte of remaining space in the tmpTokenBuf stack buffer before writing escape-processed data. There is no public exploit identified at time of analysis, and the issue is not listed in CISA KEV; it is fixed in version 3.4.1.14.
Remote code execution in Cherry Studio 1.2.2 through 1.9.12 lets attackers who control search-provider content run arbitrary Node.js code because the SearchService loads that content into an Electron BrowserWindow with nodeIntegration enabled and contextIsolation disabled. An attacker controlling a search engine provider, search result page, or provider settings page can execute JavaScript with full Node.js privileges under the OS account running the app. Publicly available exploit code exists (VulnCheck/Mundi-Xu gist) and a vendor patch (commit 1518530) is available; the flaw is not currently listed in CISA KEV.
Authenticated remote code execution in LangBot (pip package 'langbot', versions <= 4.10.5) allows any logged-in user to run arbitrary OS commands on the host by registering a malicious 'STDIO' MCP server through the Extensions > MCP configuration UI. Because LangBot passes the user-supplied command straight to the MCP SDK's StdioServerParameters, which spawns it as a subprocess, an attacker who signs up or uses stolen credentials gains full host takeover. A step-by-step and video proof-of-concept is public (publicly available exploit code exists), though there is no confirmed active exploitation.
Out-of-bounds write in the Zephyr RTOS ADIN2111/ADIN1110 Ethernet driver (versions 3.7.0-4.4.x) allows an adjacent-network attacker to corrupt up to ~14.8 KB of kernel memory by sending crafted Ethernet frames over a 10BASE-T1S or 10BASE-T1L single-pair segment. The driver's OPEN Alliance SPI frame-reassembly function copies attacker-controlled 64-byte chunks into a fixed 1524-byte static buffer without validating the write cursor against the buffer boundary, enabling writes into adjacent driver and kernel memory. SSVC indicates a proof-of-concept exists; however, EPSS is low (0.23%, 14th percentile) and CISA KEV does not list active exploitation, suggesting the flaw is known but not yet widely weaponized.
Remote code execution in PyTorch Lightning through 2.6.5 allows an attacker who can get a victim to load a malicious checkpoint file to execute arbitrary code. The flaw lives in the _load_state routine, which imports and calls the module path named in a checkpoint's _instantiator hyperparameter, letting a crafted .ckpt bypass torch's weights_only=True safeguard when LightningModule.load_from_checkpoint is invoked. Reported by VulnCheck, it is fixed in commit d710d68 (PR #21832) via an instantiator allowlist, and publicly available exploit code exists though it is not listed in CISA KEV.
Authenticated remote code execution in MantisBT versions 1.3.0 through 2.28.3 allows an administrator to run arbitrary PHP as the web server user (www-data) via the 'Manage Configuration' page (adm_config_set.php). When a config value is stored with a non-string type, the ConfigParser/Tokenizer path calls eval() on attacker-controlled code guarded only by a 'return;' prefix; because PHP hoists class and function declarations at compile time regardless of the return, an attacker can plant a class that later hijacks the autoloader. There is no public exploit identified at time of analysis and it is not listed in CISA KEV; the flaw was reported privately by watchTowr and fixed in 2.28.4.
Remote code execution in Metabase (open-source business intelligence platform) versions 1.54.0 through the fixed releases lets a user holding database-connection management rights run arbitrary code on the Metabase server. By pointing a Snowflake connection at an attacker-controlled server, a flaw in the bundled Snowflake JDBC driver writes attacker files anywhere on the host, overwriting one of Metabase's own driver files that later executes inside the Metabase process. There is no public exploit identified at time of analysis, but the vendor rates it CVSS 10.0 and a vendor security advisory (GHSA-r6x2-rchx-q9g9) confirms both the flaw and fixed versions.
Command injection in Tabby (formerly Terminus) terminal emulator before 1.0.234 lets an attacker achieve code execution by tricking a victim into dragging and dropping a crafted file whose path contains shell command-substitution metacharacters ($(...) or backticks). Because the drop handler in tabby-electron/src/pathDrop.ts inserted the raw path into the active shell without neutralizing these characters, the payload runs when the victim presses Enter. This is an incomplete-fix follow-up to CVE-2026-45038, which only addressed control characters; no public exploit is identified at time of analysis and it is not listed in CISA KEV.
Heap buffer overflow in NGINX Plus and NGINX Open Source lets unauthenticated remote attackers crash worker processes (DoS) and, on hosts with ASLR disabled or bypassed, potentially execute arbitrary code by sending crafted HTTP requests. The flaw is a data-plane-only issue triggered when a regex-based map directive references the map's regex capture variables before the map output variable in a string expression, or via a non-cacheable variable under certain conditions. F5 has released a patched version; no public exploit identified at time of analysis and it is not listed in CISA KEV.
Server-side request forgery leading to code execution in Cursor's browser-enabled Cloud Agent lets attacker-controlled web content reach an unauthenticated local agent endpoint from inside the sandbox, compromising the session. Affecting Cursor Cloud Agent sessions prior to the 03/31/2026 fix, an attacker who gets a browsing agent to load malicious content can run code in the sandbox and read repository contents, environment variables, credentials, and GitHub App access tokens scoped to that session. No public exploit identified at time of analysis, and it is not listed in CISA KEV; the flaw is a missing-authentication (CWE-306) authentication-bypass class issue with high confidentiality, integrity, and availability impact to the affected session.
Arbitrary code execution in PraisonAI (praisonaiagents) before 1.6.78 lets an attacker who can drop a Python file into a plugin directory run their own code. The plugin manager auto-discovers and executes any .py file found in project-level or user-home .praisonai/plugins/ directories at initialization, with no code signing, integrity checks, or sandboxing. VulnCheck reported it under CWE-94; there is no public exploit identified at time of analysis and it is not listed in CISA KEV.