Unauthenticated information disclosure and invitation bypass in Plane (makeplane) before 1.4.0: the GET join endpoint returns the complete ProjectMemberInvite object - invitee email, role, and acceptance token - to anonymous callers, and the companion POST accept endpoint validates only that the submitted email matches the stored invite, never the token. An attacker who knows a pending invitation's UUID can therefore learn the invited email address, register an account with it, and join the target project as the intended invitee. The primary prerequisite is possession of a non-enumerable invitation UUID, which is why complexity is rated High; no public exploit identified at time of analysis, and the issue is fixed in 1.4.0.
Stored cross-site scripting in Ghost CMS versions 5.94.0 through 6.63.x lets any staff user - including the lowest-privileged Contributor role - plant arbitrary HTML on the site's own domain through the bookmark-card feature. When a bookmark card was created, Ghost fetched the remote page's og:image and favicon URLs, trusted the URL file extension to decide the Content-Type of the stored file, and served the resulting file from the Ghost origin; an attacker-controlled source could therefore return HTML that Ghost saved and later served as an image, executing in the context of the site's domain. Because a victim staff member or administrator must subsequently open the post or editor containing the crafted card, this is an authenticated (CVSS PR:L), user-interaction-dependent stored XSS with scope change (S:C), rated 7.3; it is fixed in Ghost 6.64.0, and no public exploit code or active exploitation has been identified at time of analysis.
Stored cross-site scripting in Ghost CMS (4.22.0 through 6.64.x) allows any authenticated staff account - including the lowest-trust Contributor role - to upload unsanitized SVG media that is then served from the site's own domain, potentially compromising the admin sessions of other staff users who view it. Two upload paths are affected: SVG files supplied as video/media thumbnails, and files carrying a non-SVG extension (e.g. .png) alongside an image/svg+xml Content-Type, which the pre-patch allowlist validated only by extension. The vendor rates this 7.3, there is no CISA KEV entry and no public exploit code identified at time of analysis, and exploitation requires a higher-privileged user to actually view the malicious media (UI:R); the extension-spoofing path additionally depends on a storage adapter that serves files by content type (e.g. Amazon S3), so exposure varies by backend configuration.
Stored cross-site scripting in Ghost CMS 6.34.0 through 6.66.x allows an authenticated staff user, including a low-privilege Contributor, to store malicious scripts in post content through embed cards. The payload executes when another staff user opens the affected post in the Ghost admin editor, potentially compromising that user's admin session; it does not fire against anonymous front-end readers and requires victim interaction. The issue is fixed in Ghost 6.67.0, and no public exploit identified at time of analysis.
Stored cross-site scripting in Ghost CMS 6.22.1 through 6.63.x lets an authenticated staff user with upload privileges host a browser-executable file on the site's own domain when the default local storage adapter is in use, potentially compromising other staff members' admin sessions. The flaw arises because the local storage adapter serves uploaded files with a Content-Type derived from the file extension instead of the inert type Ghost applies elsewhere, and it requires a victim staff or admin user to load the uploaded file's URL (CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:U/C:H/I:H/A:N, base 7.3). Sites using S3 or other cloud storage adapters are not affected, and the issue is fully neutralized in 6.64.0; there is no public exploit identified at time of analysis and no confirmed active exploitation (not in CISA KEV).
Local privilege escalation in ClamXAV versions 3.3 through 3.11 allows an attacker who already holds a low-privileged local account on a macOS host to execute arbitrary code with system privileges by winning a race condition in the bundled Privileged Helper Tool that combines with insufficient file validation. Exploitation is timing-dependent (AV:L/AC:H/PR:L/UI:N in the assessed vector, with high confidentiality, integrity, and availability impact) and typically requires repeated attempts rather than a single deterministic trigger, making this a credible but not emergency-tier priority; no public exploit code has been identified at time of analysis, and ClamXAV 3.11.1 fully remediates the flaw.
Authenticated Ghost Administrators on versions 6.10.3 through 6.63.x can achieve remote code execution by installing a crafted theme that abuses how Ghost resolves theme translation (i18n) files, allowing path traversal into attacker-controlled code. Exploitation requires a high-privilege Administrator account (CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H, base 7.2) and no victim interaction beyond the attacker's own admin actions; no public exploit code has been identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV). Because the attacker population is limited to trusted or already-compromised admin accounts rather than anonymous remote users, this is a high-severity but access-gated issue rather than a mass-exploitation priority.
Authenticated code execution via media upload in Kunstmaan CMS (KunstmaanBundlesCMS) before 7.3.2: the MediaBundle checks the file-extension blacklist case-sensitively but lowercases the extension before writing the file, so a backend user with media-upload rights can submit a file named e.g. webshell.pHp, slip past the blacklist, and have it stored as webshell.php inside the web-accessible media directory. Impact escalates to full remote code execution, but only where the web server is configured to execute scripts in public/uploads/media/ - the vendor's hardening guidance (disable script execution for that path, send X-Content-Type-Options: nosniff) removes the execution impact. No public exploit code or confirmed in-the-wild exploitation was identified at time of analysis, and the flaw is gated behind a valid media-role account (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, 7.2).
Authenticated attackers with high privileges can exploit a PHP Object Injection vulnerability in Product Feed PRO for WooCommerce (versions up to 13.5.7) to potentially execute arbitrary code, delete files, or perform other malicious actions. The flaw stems from deserialization of untrusted data (CWE-502), and successful exploitation hinges on the presence of a reachable POP gadget chain in the targeted site's code; without such a chain, impact may be limited. No public exploit code or active exploitation has been identified at time of analysis, and the CVSS score is 7.2 (High).
PHP object injection in the WP Spell Check WordPress plugin, affecting all versions up to and including 12.1, allows an attacker who already holds administrator-level privileges to pass attacker-controlled data into PHP's deserialization routine and instantiate arbitrary objects. Successful object injection can lead to high-integrity and high-confidentiality impact - including code execution, file manipulation, or privilege abuse - depending on whether a usable POP gadget chain exists in the plugin or the surrounding WordPress environment, which the available data does not confirm. With a CVSS 3.1 base score of 7.2 (High) and PR:H in the vector, the practical attack surface is limited to trusted high-privilege users; no public exploit identified at time of analysis and no vendor-released patch identified at time of analysis.
PHP object injection in the WordPress plugin WP Ultimate Exporter (Smackcoders Inc.) affects all versions up to and including 3.0, allowing an attacker who already holds a high-privilege authenticated account on the target site to submit crafted serialized PHP data to a plugin endpoint and instantiate arbitrary objects. Successful exploitation depends on a usable POP gadget chain being present in the plugin, WordPress core, or a loaded library; without one, the flaw may only yield object injection rather than full code execution. No public exploit code or confirmed active exploitation was identified at time of analysis, and the principal barrier is the CVSS PR:H prerequisite, which limits realistic abuse to privileged insiders or attackers who have already compromised an administrative account.
PHP object injection in the WordPress plugin Sunshine Photo Cart (all versions through 3.7.1) allows an attacker who already holds a high-privilege authenticated WordPress account to submit crafted serialized data to a plugin endpoint, potentially achieving object injection with high confidentiality, integrity and availability impact. The CVSS 3.1 vector (AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H) rates this 7.2 and confirms the attacker must be authenticated with elevated privileges (PR:H), which excludes anonymous internet users and is the primary limiting factor; no user interaction is required. No public exploit code has been identified at time of analysis, and there is no confirmed evidence of active exploitation (not listed in CISA KEV) - realistic impact depends on a usable PHP gadget chain being present in the plugin or the hosting stack, which is not confirmed by the available data.
An out-of-bounds write in Android's wpa_supplicant Robust AV / SCS action-frame handler (wpas_handle_robust_av_scs_recv_action in robust_av.c) can be turned into remote code execution on Android 14 through 17 (including 16-qpr2) devices that receive a crafted SCS action frame. Exploitation is sharply constrained: per the CVSS vector PR:H, the attacker must already hold System-level execution privileges on the target before the malformed frame is processed, and the independently assessed vector (AV:A) places the attacker in Wi-Fi/adjacent range rather than on arbitrary internet-routable paths. User interaction is not required (UI:N), and no public exploit code or CISA KEV entry has been identified at time of analysis; combined with the vendor score of 7.2 and the adjacent-vector reassessment, this rates as a moderate, not emergency, priority.
Incorrect access control in Feehi CMS 2.1.1 allows a low-privilege backend administrator who holds the administrator-update permission to overwrite the password of the built-in super administrator account, achieving full super-admin takeover. The server neither protects the built-in super administrator record from modification nor requires the target account's current password during the update flow, so a single crafted backend request is sufficient. This is an authenticated privilege-escalation issue (CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H, rated 7.2 High) that requires a pre-existing privileged backend session; no public exploit code or confirmed active exploitation was identified at time of analysis, and no vendor-released patch was identified in the available intelligence.
Authenticated workspace members can tamper with another application's MCP server configuration in Dify versions prior to 1.16.0 by sending a PUT request to /console/api/apps/<app_id>/server with an arbitrary server ID, allowing them to disable or redirect the targeted app's MCP integration. The flaw is a cross-tenant/IDOR access-control gap (CWE-639) rated CVSS 7.1 (PR:L, I:H, A:L) that matters most in multi-tenant or multi-team Dify installations; no public exploit identified at time of analysis.
Cross-tenant file disclosure in Langflow 1.6.8 through 1.9.0 allows any authenticated user who can reach a project-scoped MCP connection to read another user's flow-backed artifacts - uploaded documents, structured data, prompts, and other private flow files - by supplying an arbitrary resource URI to the resources/read handler, which forwarded the caller-controlled URI to storage_service.get_file without verifying that the referenced flow belonged to the caller or the current project (CWE-639). The flaw is read-only, so victim files and stored flows are not modified, but confidentiality impact is high on multi-tenant or multi-user instances where other users' files exist to be read; no public exploit code identified at time of analysis and the issue is not confirmed as actively exploited (not in CISA KEV). The vendor fixed the issue in Langflow 1.9.1.
Cross-tenant deletion of estimate points in Plane (Makeplane) before 1.4.0 allows any authenticated user holding an administrator or member role in one workspace to permanently destroy an estimate point owned by a different workspace, simply by supplying the victim record's UUID in a URL scoped to the attacker's own workspace. The flaw is a destructive cross-tenant IDOR (CWE-639) rated CVSS 7.1 (AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L), requiring only a valid low-privilege account and knowledge of the target's random 128-bit UUID, which must be obtained out-of-band since it is not enumerable through this endpoint. No public exploit code or confirmed active exploitation was identified at time of analysis; the operational risk is moderate-to-high specifically for multi-tenant Plane deployments, while single-tenant installs have no cross-tenant victim.
Denial of service in Moby BuildKit allows a low-privileged remote user to crash the BuildKit daemon by submitting a build that attempts to use a CDI device when the daemon was started with the --cdi-disabled flag. This triggers a null pointer dereference (CWE-476) and daemon panic. The vulnerability requires a non-default configuration and the ability to submit builds; default deployments are not affected. No public exploit has been identified, and the vendor has released a patch in version 0.33.1.
Authenticated low-privilege users can retrieve embedded sensitive data from WordPress sites running the RepairBuddy plugin (versions through 4.1226) due to a missing authorization check. This moderate-severity vulnerability (CVSS 7.1) impacts only confidentiality, requires a valid account with minimal privileges, and has no public exploit code or active exploitation reported.
vLLM deployments prior to 0.30.0 with prefix caching enabled (the default) fail to carry the caller's cache_salt value across Harmony tool continuations submitted to "POST /v1/responses", so continuation prefixes land in the global unsalted prefix-cache namespace. This lets an authenticated tenant who can reconstruct a victim's low-entropy post-tool conversation history submit the same continuation and read the cached_tokens_per_turn counter as a hit/miss oracle, confirming whether that prefix was previously processed and thereby defeating salt-based tenant isolation of the cache. This is a genuine but low-severity, low-priority issue: the assessed vector is CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:L/I:N/A:N, exploitation requires authenticated access, prefix caching to be on, cache_salt salting to be in use, and the victim's history to be low-entropy and reconstructable, and only the isolation property leaks - no direct model output or prompt content is disclosed. No public exploit code and no confirmed active exploitation were identified at time of analysis.
Authenticated remote users of Langflow can defeat the localhost-only restriction on the MCP configuration install endpoint (POST /api/v1/mcp/project/{project_id}/install) by sending a forged 'X-Forwarded-For: 127.0.0.1' header, causing the server to treat the request as local and allowing the attacker to write or overwrite an MCP client configuration file on the host filesystem. The flaw affects Langflow versions 1.5.0 up to but not including 1.10.3 and carries a CVSS v3.1 score of 7.1 (AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L), reflecting a high integrity impact and low availability impact but no confidentiality loss. Authentication is mandatory, no user interaction is required, and no public exploit identified at time of analysis; no EPSS data was supplied for this CVE, so risk is driven by the moderate CVSS score and the requirement for a valid session rather than by observed exploitation.
Reflected cross-site scripting (XSS) in the VillaTheme 'Photo Reviews for WooCommerce' plugin for WordPress affects all versions through 1.2.30, allowing an unauthenticated attacker to inject script into a page viewed by a victim. Exploitation is phishing-driven: the attacker must lure a targeted visitor to a crafted URL, and the payload executes only in that victim's browser session, giving CVSS 3.1 a base score of 7.1 (AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:L) rather than server-wide compromise. No public exploit code identified at time of analysis, and the issue is not listed as confirmed actively exploited (CISA KEV); the vulnerability was reported by Patchstack and the exact vulnerable parameter or endpoint is not detailed in the available data.
Reflected cross-site scripting (XSS) in the Adsmonetizer WordPress plugin (adsensei-b30) through version 3.2.4 allows remote attackers to inject arbitrary script into a victim's browser by convincing them to click a crafted link. The attacker does not need authentication, but exploitation requires user interaction and the targeted site must have the vulnerable plugin installed and reachable. This is rated CVSS 7.1, and no public exploit has been identified at time of analysis; it is not listed as confirmed actively exploited (CISA KEV).
Cross-Site Request Forgery (CSRF) in the Blubrry PowerPress Podcasting plugin for WordPress up to version 11.17.9 allows remote attackers to force a logged-in user, typically an administrator, to perform unintended actions by tricking them into interacting with a crafted link or malicious page, due to missing CSRF token validation. Exploitation requires user interaction (UI:R) and an active authenticated session, but no public exploit code is currently identified. The vulnerability has a CVSS score of 7.1 (moderate severity per independent assessment) and is tracked as CWE-352.
An insecure direct object reference (IDOR) in Plane prior to 1.4.0 lets an authenticated member of a project reassign ownership of another user's issue attachment by supplying a pk UUID in the V2 issue-attachment PATCH request. Because the endpoint validates only pk, workspace, and project_id while ignoring the issue_id present in the URL, an attacker with project membership (CVSS PR:L, network vector) can target any attachment still in the PENDING, not-yet-confirmed upload state and set created_by to their own account, thereby taking control of it; confirmed attachments are unaffected, and exploitation also requires knowing the victim's unguessable UUID. The flaw is a genuine but moderate-priority authorization issue (CWE-639) with no public exploit identified at time of analysis, and it is fixed in the vendor's 1.4.0 release.
Insecure direct object reference in Plane's public Spaces asset endpoints (all versions prior to 1.4.0) allows an attacker to read, overwrite, or restore arbitrary FileAsset records belonging to other projects and workspaces by supplying a crafted asset ID. The vulnerable code paths - EntityAssetEndpoint and AssetRestoreEndpoint in plane/space/views/asset.py - resolve a DeployBoard from a public anchor but fetch or mutate FileAsset rows scoped only to that board's workspace, with no membership check and no project_id constraint, even though sibling endpoints in plane/app/views/asset/v2.py had already been hardened for two earlier cross-tenant asset IDORs tracked as CVE-2026-27705 and CVE-2026-46558. Exploitation requires that at least one board be published as a public/shared board so an attacker can obtain a valid DeployBoard public anchor; the input CVSS 3.1 vector rates PR:L (7.1, High), while our independent assessment scores PR:N because the affected endpoints are reachable under the unauthenticated /api/public/ namespace, and this difference should be confirmed in the target deployment. No public exploit code or in-the-wild exploitation has been identified for this issue at time of analysis, and it is not listed in CISA KEV.
Code execution in Newell Brands DYMO ID 1.5.1.71 occurs because the application resolves its plugin Modules directory relative to the process working directory, letting an attacker who places a malicious plugin module or DLL alongside a crafted job file load arbitrary code at the victim's privilege level once that job file is opened. The issue is local and file-delivery based rather than network-reachable, requires explicit victim interaction (opening the crafted job), and grants only the privileges the victim already holds, so it cannot self-propagate; no public exploit code and no CISA KEV entry were identified at time of analysis. The vendor has released a fix in DYMO ID 1.6.0.
Code execution via local privilege escalation is possible in Google Android's Bluetooth AVRCP handling (handle_app_val_response in btif_rc.cc) because of a race condition (CWE-362) that can corrupt shared state in the Bluetooth stack. The flaw is rated CVSS 7.0 (AV:L/AC:H/PR:L/UI:N) and affects Android 16, Android 16-qpr2, and Android 17; an attacker must already have a low-privilege app installed on the device and must repeatedly drive the AVRCP remote-control code path to win a narrow timing window, with no user interaction required. No public exploit code has been identified and CISA's SSVC assessment records no observed exploitation, so while the technical impact is total if the race is won, the high attack complexity and local-only prerequisites keep this a lower-urgency item than a remotely reachable, deterministic bug.
Unauthenticated guests can force execution of any published automation task in TF Content for Joomla versions 2.9.0 through 2.9.4 by supplying a task's numeric ID to the records.custom_action endpoint. The extension lacks authentication, ACL, CSRF, task-trigger, content-binding, and cron-token enforcement, allowing immediate dispatch of the configured executor. No public exploit identified at time of analysis, and the risk is assessed as a genuine but mid-tier priority.
Local file disclosure in Docling's OpenDocument backend (versions 2.107.0 through 2.120.3) allows an attacker-supplied ODF file to force the parser to read arbitrary filesystem paths. When a draw:image element's xlink:href references a part missing from the document archive, the vulnerable backend treats the href as a local filesystem path without a scheme check or extraction-directory confinement, then embeds Pillow-decodable files into the converted output; non-image paths still leak existence information through read success or failure. Exploitation requires a victim or automated pipeline to convert a malicious ODF document (UI:A) in an unauthenticated network-reachable context (PR:N), and no public exploit code or CISA KEV active-exploitation entry has been identified at time of analysis. Note that on affected versions this code path bypassed the enable_local_fetch setting that other Docling backends honor, so the read occurred regardless of that configuration. The vulnerability is fixed in 2.120.3.
Information disclosure and entitlement bypass in Docker Buildx Bake allows an attacker-supplied Bake definition to read files that should have required an explicit fs.read entitlement approval. An unauthenticated attacker can craft a definition that exposes a readable file via a pathless secret whose ID is interpreted as a client-side pathname, or that consumes a local OCI image layout outside the project because entitlement validation checks a different path representation than the one actually used. Exploitation requires the victim to run an untrusted Bake definition (CVSS 4.0 AV:N/AC:L/PR:N/UI:A, assessed as CVSS 3.1 AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:N), impact is read-only with no code execution, write, or availability loss, and no public exploit has been identified at time of analysis; environments that build only their own trusted first-party Bake definitions are not exposed. This is a genuine but moderate-priority entitlement-bypass issue rather than a critical remote threat.
Authorization bypass in CozyThemes Cozy Blocks WordPress plugin versions up to 2.2.23 allows unauthenticated remote attackers to access data they are not authorized to view via an insecure direct object reference (IDOR). The flaw, rated CVSS 4.0 6.9 by Patchstack, requires no authentication or user interaction and results in low confidentiality impact only. No public exploit code has been identified at time of analysis.
TF Content 2.9.0 through 2.9.4, a Joomla content extension distributed by joomlafry.com, allows unauthenticated remote attackers to create and modify arbitrary TF Content records because its public RecordController performs no authorization check and its shared frontend save controller writes the raw jform array without filtering it through the configured form. By pulling a valid Joomla token from the public login form and supplying an existing record ID, a guest - with no account, privileges, or user interaction (CVSS:3.1/AV:N/AC:L/PR:N/UI:N) - can mass-assign fields such as published, access, and created_by on any TF Content row. Impact is bounded to low confidentiality, integrity, and availability effects on TF Content data rather than full site compromise; no public exploit code or confirmed active exploitation was identified at time of analysis, placing this as a genuine but moderate-priority issue.
Authorization bypass in Themeisle's AI Chatbot for WordPress - Hyve Lite plugin (versions through 2.0.2) allows unauthenticated remote attackers to modify chatbot-related records they do not own by supplying a user-controlled object identifier, per the assessed CVSS:3.1 vector AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N. This is a moderate-severity IDOR (CWE-639) tracked as CVE-2026-97305 and scored CVSS 4.0 6.9; exploitation is limited to a Low integrity impact, with no confidentiality or availability effect, and no public exploit was identified at time of analysis.
Information disclosure in the Zabbix server's JavaScript preprocessing engine (embedded Duktape interpreter) allows an authenticated 'limited administrator' account to read raw heap memory and potentially expose data belonging to other concurrently running preprocessors that the account should not be able to access. The root cause is an out-of-bounds read (CWE-125) in the JavaScript preprocessing path; exploitation requires only the ability to define or run a JavaScript preprocessing step, and there is no integrity or availability impact. No public exploit code was identified at time of analysis and there is no CISA KEV entry; this is a moderate-severity privilege-boundary/tenant-isolation issue (CVSS 4.0 base 6.9; assessed CVSS 3.1 AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:N/A:N) rather than an unauthenticated remote compromise.
Unauthenticated information disclosure in GROWI (CVSS 5.3) lets a network attacker read files attached to non-public pages when the instance's file upload setting is configured as 'Local' rather than a cloud or object-storage backend. Exploitation requires no credentials and no user interaction (PR:N/UI:N), but the impact is limited to low-confidentiality exposure of files that already reside on the server, with no write access or code execution. No public exploit code has been identified at time of analysis and the flaw is not reported as confirmed actively exploited (CISA KEV); the 'Local' storage precondition is the primary factor bounding the exposed population.
Unauthenticated remote credential disclosure in SGLang (sgl-project/sglang) up to version 0.5.21, where the /server_info HTTP endpoint and the gRPC GetServerInfo path returned the entire ServerArgs dataclass and thereby exposed the configured api_key, admin_api_key, and ssl_keyfile_password values verbatim to any network-reachable caller. No privileges or user interaction are required (CVSS 4.0 vector PR:N/UI:N, with E:P indicating a proof-of-concept), and publicly available exploit code exists; KEV has not flagged this as actively exploited. Impact is confidentiality-only but potentially severe - a leaked admin API key gives full control of the inference service - with the practical gate being whether the operator actually configured a secret and left the management endpoint exposed. A vendor fix exists only as an unmerged pull request (PR #30343), so released patched versions could not be confirmed at analysis time.
Zabbix Server and Zabbix Proxy accept the active-agent heartbeat message without validating the configured PSK or TLS certificate, so anyone who can reach the trapper port (default TCP 10051) and knows or guesses a valid host identifier can report an arbitrary host running an active agent as available. The resulting impact is integrity-only - falsified host availability state - with no data exposure, no code execution and no denial of service; our independent assessment (CVSS 3.1: AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N) matches the vendor's CVSS 4.0 base score of 6.9 in treating this as a moderate, non-emergency issue. No public exploit code or confirmed active exploitation was identified at time of analysis.
A malicious frontend can crash the BuildKit daemon (buildkitd) by submitting a crafted LLB (Low-Level Build) definition, causing a panic that terminates the daemon and interrupts all running builds. The vulnerability affects BuildKit versions prior to v0.33.1 and requires the attacker to have access to the build submission interface; no authentication is needed (PR:N) and no user interaction is required (UI:N), but the attack vector is local/adjacent (AV:L), making shared or multi-tenant build infrastructure the primary risk. No public exploit code has been identified at the time of analysis, and the vendor has released a patch in v0.33.1.
Denial of service in Moby BuildKit allows anyone able to submit an LLB build definition to a buildkitd daemon to panic and terminate the daemon process, aborting every build running on that instance. The flaw is an out-of-bounds array index (CWE-129) reached during processing of an attacker-controlled low-level build graph, and it affects all BuildKit releases prior to v0.33.1, including BuildKit embedded in Docker Engine and CI/build-service deployments. The assessed impact is availability-only (CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, 6.9), the attack surface is the local build API/daemon socket rather than arbitrary network exposure, and severity scales with multi-tenancy - a shared CI runner or multi-tenant build service loses all concurrent builds from a single crafted definition. Vendor patch v0.33.1 is available; no public exploit code was identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV).
Stored cross-site scripting in Ghost CMS versions 4.0.0 through 6.66.x allows an attacker to plant malicious JavaScript on the victim site's own domain by convincing an Administrator to import a crafted content archive containing an unsanitized SVG image. Per the CVSS vector (AV:N/AC:H/PR:N/UI:R), the attacker does not need credentials but the attack is not self-service: it depends on successful social engineering of a privileged Administrator to run the import (UI:R), and impact on staff users' admin sessions further depends on that stored SVG being rendered in an authenticated admin context. Exploitation is therefore conditional rather than opportunistic, and no public exploit code has been identified at time of analysis (no CISA KEV listing).
Local privilege escalation in MediaTek's meta component affects numerous MediaTek chipsets (including MT6789, MT6895, MT6985, MT8195, and others listed in EUVD-2026-92196) and allows an attacker with physical access to the device to trigger an out-of-bounds write due to a missing bounds check, achieving high confidentiality and integrity impact. No authentication, existing execution privileges, or user interaction are required once physical access is obtained. No public exploit has been identified at time of analysis; the physical access prerequisite limits remote exploitation and confines realistic abuse to scenarios such as lost or stolen devices, border inspection, or supply-chain tampering.
Physical access to a Mercusys AC12 V2 router's printed circuit board lets an attacker take full control of the device: the UART debug header on the board is exposed with no authentication gate or lockdown, so wiring a serial adapter to the correct pins yields an interactive, root-level console from which arbitrary code can be executed (CWE-1299, unprotected alternate hardware interface). The CVSS 3.1 base score of 6.8 (AV:P/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) reflects complete loss of confidentiality, integrity, and availability, tempered only by the requirement for hands-on hardware access - there is no network, Wi-Fi, or LAN attack path, so remote exploitation is not possible. No confirmed active exploitation and no public exploit code have been reported, and the EPSS score is 0.16% (4th percentile), consistent with a flaw whose realistic exposure is limited to supply-chain, device-resale, lost/stolen, or evil-maid scenarios.
LibreOffice Calc fails to confine file operations when processing an external data-source link to an embedded Firebird database, allowing a crafted document to write files to any location the user has write permission to. Exploitation requires the victim to open the malicious spreadsheet (user interaction) and does not by itself escalate privileges; no public exploit or active exploitation has been identified at time of analysis. The vulnerability is a moderate-severity, user-interaction-dependent file-write issue (CVSS 6.8) rather than a high-priority remote threat.
Improper validation of integrity check values in HP ThinPro thin-client OS images allows an attacker who already has physical access to a device to supply tampered data that the system fails to verify, resulting in high confidentiality and integrity impact (CVSS:3.1 AV:P/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N; vendor CVSS 4.0 base 6.8). Organizations are affected if they run HP ThinPro 8.1 prior to Service Pack 10 or HP ThinPro 9 prior to Service Pack 3, typically on kiosks, lobbies, and shared VDI endpoints where physical access is loosely controlled. There is no public exploit identified at time of analysis and the issue is not listed in CISA KEV, so the realistic risk is moderate and tightly constrained by the mandatory physical-access prerequisite rather than by remote exposure.
Denial of service in the BuildKit Dockerfile frontend (buildkitd) allows a user who can submit a build context to exhaust the daemon's memory and terminate it by placing an oversized Dockerfile or .dockerignore file in that context, because these files were loaded into memory without any size limit (CWE-789). The assessed vector (AV:L/PR:L) indicates low-privilege, local/context-level access is required to submit the build, so the practical risk is concentrated on shared or multi-tenant buildkitd instances such as CI/CD pipelines and hosted build services, where one submitter's crash interrupts other users' builds, while a single-user workstation daemon only harms its own environment. Impact is bounded to availability with no code execution, data disclosure, or integrity impact, no public exploit code or CISA KEV confirmation was identified at time of analysis, and the vendor has released BuildKit v0.33.1, which rejects such files above 16 MiB.
Cross-workspace project modification in Plane prior to 1.4.0 allows an authenticated administrator of one workspace to alter project metadata and configuration in another workspace by supplying a known target project UUID. The vulnerability stems from an IDOR (CWE-639) in the project partial-update endpoint, which validates the caller against the workspace slug in the URL but loads the project globally without binding it to that workspace. Exploitation requires admin privileges and knowledge of the victim project's UUID, and impact is limited to multi-tenant deployments; no public exploit identified at time of analysis.
Opening a crafted LibreOffice Calc document can cause a local text file on the victim's machine to be read into the spreadsheet, leaking its contents to anyone who later views or receives the file. The document must embed a saved external data link of the sql type that points at a local folder of text files as a database; vulnerable versions automatically restore that link at load time, whereas fixed versions restore only the csv, html and xml data providers. Exploitation requires the victim to open the attacker-supplied document (user interaction, UI:P in the vendor's CVSS 4.0 vector of 6.7, and UI:R in the independent CVSS 3.1 assessment), so exposure is limited to targeted social-engineering scenarios where the attacker also knows or guesses a useful local path and has a way to retrieve the populated sheet; no public exploit identified at time of analysis and the issue is not in CISA KEV.
An out-of-bounds read/write flaw in MediaTek's ccci (cross-core/modem control) component allows an authenticated local attacker who already holds System privilege to corrupt memory, potentially leading to information disclosure, crashes, or further privilege escalation. Exploitation requires local access and, per the vendor description, user interaction - though the vendor CVSS vector omits UI:R - and is limited to specific MediaTek chipsets (MT6880, MT6890, MT6980D, MT6990, MT6986, MT6986D, MT6813, MT6988, MT2735, MT2737). No public exploit code has been identified, SSVC reports exploitation: none and not automatable, making this a high-severity but low-urgency issue due to its high-privilege prerequisite.
Environment-variable and INI-file value disclosure in LibreOffice lets a crafted document exfiltrate host configuration data to a remote server when the victim opens it, because document-supplied URLs can still be expanded in code paths that the earlier CVE-2024-12426 fix missed. The incomplete fix overlooked XForms instance data and the Calc CSV and SQL data providers, which continue to resolve internal-scheme URLs originating from the document. This is a moderate-severity, confidentiality-only flaw (CVSS 4.0 base 6.7, CWE-200) that requires the victim to open an attacker-supplied file, with no public exploit identified at time of analysis; the independent assessment scores it as a local-vector, passive-user-interaction issue with no impact on integrity or availability.