Authentication bypass in Grav Login plugin versions 2.7.0 before 3.9.9 allows attackers who already know a victim's password to disable two-factor authentication (2FA) and take over the account by invoking the update_user task through a published page form without an access rule. The flaw resides in processUserProfile(), which verifies that a session user exists but fails to confirm that a pending front-end 2FA session is fully authorized. Exploitation requires three conditions: the attacker must possess the victim's password, have an in-progress front-end login that has reached the 2FA step, and be able to publish a Grav page form lacking an access rule; no public exploit has been identified at time of analysis.
Improper privilege management in Rapyd Admin (the open-source Rapyd Admin scaffolding package for Laravel) versions 9.0.0 through 9.19.x lets an authenticated user who already holds the 'edit users' permission - the shipped 'operator' role is sufficient - promote their own account to the 'admin' role by submitting the admin role ID through the UsersEdit Livewire component, which prior to 9.20.0 synced browser-supplied roles without validating them against a server-side allow-list. Successful exploitation lifts tenant scoping, bypasses per-record authorization and enables impersonation of other users, including higher-privileged accounts, giving high confidentiality, integrity and availability impact within the affected installation. The flaw is fixed in v9.20.0, which computes assignable roles server-side and prevents a non-super-admin from granting a role to itself; no public exploit code and no confirmed active exploitation were identified at time of analysis, and the prerequisite of an already-delegated privileged account makes this an insider or compromised-account escalation rather than a default-configuration, unauthenticated issue.
Configuration-gate bypass in Crawl4AI 0.9.0 through 0.9.3 lets an authenticated but non-admin API client submit a laundered LLMConfig object through the server's configuration endpoint and read back environment-resolved secrets, including SECRET_KEY and provider API keys, which can then be used to forge admin JWTs. Exploitation requires Crawl4AI to be running in server/API mode with the /config/dump endpoint reachable and the attacker to already hold a provisioned non-admin API token (CVSS 4.0 base 8.6, vector PR:L), so pure library or embedded deployments are not affected; impact is highest where SECRET_KEY and provider keys are present in the server environment. Independent assessment flags this as a credible high-priority issue for hosted Crawl4AI deployments, with no public exploit identified at time of analysis and no CISA KEV listing.
Authorization bypass in @neo4j/graphql 7.0.0 through 7.6.3 lets an authenticated user read, create, and delete records that @authorization filter rules were written to deny, because the library translates the 'none' quantifier on relationship and connection filters into Cypher without the required negation so the predicate is evaluated as though it were 'some'. The flaw only manifests in applications whose typedefs actually use the 'none' quantifier inside an @authorization filter on a relationship or connection field (for example creator: { none: { id: { eq: "$jwt.sub" } } }), so applications that never use that construct are not affected; the CVSS vector specifies PR:L, meaning the attacker must hold a valid JWT accepted by the API. No confirmed active exploitation (not in CISA KEV) and no public exploit code identified at time of analysis; a vendor patch and the upstream fix commit are available.
Reflected cross-site scripting in Frappe ERPNext (all versions before 14.92.1, and 15.x before 15.83.0) lets an attacker inject arbitrary JavaScript that runs inside an authenticated user's session, enabling session/account takeover. The attacker needs no credentials (PR:N in the CVSS vectors), but exploitation requires a logged-in victim to follow a crafted URL (UI:P), so it is delivered by phishing or social engineering rather than by a direct server-side attack. Impact is rated high for confidentiality and integrity (VC:H/VI:H) with scope change under the CVSS 4.0 vector (8.6) and CVSS 3.1 equivalent (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:N); there is no CISA KEV entry and no public exploit identified at time of analysis, and the issue is lower priority than a zero-interaction network RCE.
SurrealDB versions before 3.3.0 fail to validate a caller's authorization level when the SurrealQL USE statement switches the session to a different namespace or database, allowing an authenticated tenant to cross tenant boundaries and read, create, update, or delete rows in other tenants' tables wherever the table's PERMISSIONS clause permits it (for example PERMISSIONS FULL). Exploitation requires a valid account at the namespace, database, or record level (CVSS 4.0 vector PR:L, unauthenticated access is not possible) and only has impact on instances hosting multiple tenants with permissively configured tables. The vendor has released 3.3.0 to correct the missing authorization check; no public exploit code has been identified at time of analysis, and the flaw is not listed in CISA KEV.
Cross-tenant data access in SurrealDB before 3.3.0 allows an authenticated namespace- or database-level SYSTEM user to bypass namespace/database scoping and read, create, modify, or delete rows and read DEFINE PARAM values in other tenants where table PERMISSIONS clauses permit. The CWE-639 authorization bypass is triggered by setting the Surreal-NS/Surreal-DB headers on the /sql endpoint or invoking the RPC 'use' method, as the server fails to validate the caller-selected namespace or database against the level at which the user authenticated. No public exploit identified at time of analysis; impact is limited to multi-tenant deployments and requires valid low-privilege credentials.
OS command injection in mac-shell-mcp before 2.0.0 lets a connected MCP client - including a legitimate client driven by a malicious or prompt-injected LLM - escape the tool's command whitelist by appending shell metacharacters to the arguments of an otherwise permitted command such as pwd, causing forbidden commands to run as the server user without approval. Exploitation requires an existing MCP client session to the local server (CVSS 4.0 vector AV:L/PR:L/UI:N, mirroring the independent CVSS 3.1 assessment AV:L/PR:L/UI:N with high confidentiality, integrity and availability impact), so this is an authenticated, locally-scoped flaw rather than an internet-facing unauthenticated one, and impact is bounded by the privileges of the account the server runs under. No public exploit code has been identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV), but a vendor fix and upstream patch commit are available; EPSS data was not provided for this CVE.
Stored cross-site scripting in Frappe ERPNext (before 15.119.1, and 16.x before 16.32.0) allows unauthenticated visitors to plant crafted HTML/script through a public-facing booking page; the payload is stored without neutralization and executes in the browser session of a privileged user who later views the booking content, enabling account takeover of that user. Exploitation is conditional: the instance must expose the public booking feature to untrusted submitters, and impact only completes when an authenticated, higher-privileged user opens the tainted content (UI:P in the CVSS 4.0 vector), so the attack is not fully automatable to immediate compromise and no public exploit code was identified at time of analysis. The vendor has released fixed versions, and this is a genuine priority for organizations that run the vulnerable booking workflow, not merely a high-score-on-paper issue.
Privilege escalation in Progress MarkLogic Server 12.x before 12.1.0 and 11.x before 11.3.7 allows an attacker who can supply user-controlled values to certain higher-order XQuery functions to have those values evaluated in an elevated security context, potentially gaining permissions beyond those granted. The vendor's CVSS 3.1 vector (8.6 High, AV:N/AC:L/PR:N/UI:N/S:C/C:N/I:H/A:N) claims unauthenticated exploitation, but the independent assessment assesses PR:L and high confidentiality and integrity impact, consistent with the requirement for access to an XQuery-evaluating interface (a MarkLogic account with query/evaluation rights or an application that passes user input into such functions). No public exploit code or CISA KEV entry is present in the provided intelligence, and no EPSS score was supplied.
Local code execution in Samsung's HEIF decoder library (libsimba.heifdec.media.samsung.so) allows an unprivileged local attacker to execute arbitrary code by supplying a crafted HEIF image on devices running firmware prior to SMR Oct-2026 Release 1. The flaw is an improper input validation issue in the 'sheifd' parsing path, rated 8.6 (CVSS 4.0) with high confidentiality, integrity, and availability impact. No public exploit code has been identified at the time of analysis, and the local attack vector requires the malicious file to be delivered from an on-device context.
Local code execution in Samsung's HEIF decoder library (libsimba.heifdec.media.samsung.so) allows unprivileged attackers to run arbitrary code by supplying a crafted HEIF cover image. This improper input validation flaw, tracked as CVE-2026-21124, affects Samsung devices prior to the SMR Oct-2026 Release 1 security update and carries a CVSS 4.0 score of 8.6, indicating high confidentiality, integrity, and availability impact. No public exploit code has been identified at time of analysis, and patched devices are not affected.
Session token theft in Open WebUI 0.7.0 through 0.11.3 lets any website a signed-in user visits post an unvalidated 'verify:chat' message that the SyncStatsModal component accepts without checking the sender's origin, ultimately returning the victim's JWT and granting full API access to their account. The attack is client-side and remote, scoring 8.6 (CVSS 4.0) and independently assessed at CVSS 3.1 8.x severity (AV:N/AC:L/PR:N/UI:R, C:H/I:H); it requires no server-side authentication, but does require that the victim hold a live session and interact with an attacker-controlled page. No public exploit code or active exploitation has been identified at time of analysis, and the vendor has released a fix in 0.11.4.
Stored cross-site scripting in EdgeEver 1.6.0 through 1.112.0 allows an attacker with a valid API or MCP token to inject HTML event handlers into template contentMarkdown, which executes when a victim opens the template preview, granting access to the privileged window.edgeeverDesktop bridge. Exploitation requires a low-privileged authenticated token (PR:L) and user interaction to open the preview (UI:P), and no public exploit code has been identified at time of analysis.
Path traversal in the legacy InfoStudio modules shipped with Progress MarkLogic Server lets a low-privileged authenticated remote attacker reach files outside the intended module directory and, because user-context values are also insufficiently validated, potentially have code run under an unintended identity. Affected deployments are MarkLogic Server 12.x releases before 12.1.0 and 11.x releases before 11.3.7 that still have the legacy InfoStudio modules present and reachable. The vendor rates this 8.5 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N), reflecting a confidentiality-first impact that crosses a security boundary; no public exploit code has been identified at time of analysis and the flaw is not confirmed as actively exploited, but fixed builds are already available, so this is a priority patching item rather than an emergency.
Authenticated role administrators in Admidio 4.3.2 through 4.3.17 and 5.0.0 before 5.0.15 can add users to roles belonging to a different organization on the same instance, escaping their own tenant's authorization boundary and gaining privileged membership that lets them sign in to the target organization and modify its user profiles. The flaw is a missing cross-organization ownership check on the role UUID submitted to members_assignment.php (CWE-639), rated CVSS 4.0 8.5 and assessed independently as CVSS 3.1 8.5 (AV:N/AC:L/PR:H/UI:N/S:C/C:H/I:H/A:N). Exploitation is gated on the attacker already holding role-administrator rights and on the deployment hosting more than one organization, so single-org installations and anonymous or low-privilege users cannot reach the vulnerable path; no public exploit code has been identified at time of analysis and there is no CISA KEV listing. This is a genuine but conditional priority, bounded to insiders or compromised administrator accounts.
Password-reset token disclosure in the Grav Login plugin (versions 1.2.0 up to but not including 3.9.9) allows any user holding a publisher-level account to hijack more privileged accounts. A publisher authors a page at the plugin's password-reset route so that Login::getPage() renders attacker-controlled content on that route; when a higher-privileged victim opens an emailed reset link, the rendered page exposes the victim's username and bearer reset token, which the attacker can redeem to take over the account. Exploitation requires an existing publisher account plus victim interaction (clicking the emailed reset link), and no public exploit code or CISA KEV entry has been identified at time of analysis.
Stored cross-site scripting in Open WebUI versions from 0.6.33 up to (but not including) 0.11.4 lets an authenticated user embed a javascript: URL in a citation's embed_url field inside a chat, then share that chat so that when the victim clicks the citation the URL is handed to window.open. Successful exploitation executes script in the victim's browser session, allowing theft of localStorage.token and full account takeover as the victim. The attacker must already hold a valid account (PR:L) and the victim must actively click the crafted citation (UI:P), so exposure is limited to instances where chat sharing is enabled and users interact with untrusted shared content; no public exploit code or CISA KEV entry was identified at time of analysis, and the vendor fix is released in 0.11.4.
Better Auth versions 1.3.8-beta.3 to before 1.7.0-rc.3 with the deviceAuthorization plugin enabled are vulnerable to a UI misrepresentation flaw (CWE-451) in which the GET /device approval screen omits the requesting client_id and scope, letting unauthenticated remote attackers initiate a device code and trick a signed-in user into approving it, resulting in a live session token. Exploitation requires user interaction and is not actively exploited; no public exploit identified at time of analysis, and the independent assessment rates it as a credible but not critical-severity risk.
Stored cross-site scripting in Frappe ERPNext allows an authenticated user with permission to create or rename records to plant script payloads in record names that are later rendered without proper neutralization on certain dashboards, tools, and portal pages. Versions before 15.103.0 and 16.x before 16.12.0 are affected; the vulnerability scores CVSS 4.0 8.5 (AV:N/AC:L/AT:N/PR:L/UI:P/VC:H/VI:H/VA:N) and the independent assessment rates it a credible, above-average-priority issue rather than an emergency, because a valid low-privilege login is mandatory, the payload only fires when a victim - ideally a privileged user - opens the specific affected view, and the attacker cannot control exact timing. Successful exploitation can read or modify data and steal session tokens from other users' browsers; no public exploit code or confirmed active exploitation (CISA KEV) was identified at time of analysis. Installations running 15.103.0+ or 16.12.0+ are not vulnerable.
Stored cross-site scripting in Frappe ERPNext (before 15.113.0 and 16.x before 16.24.0) allows an authenticated low-privilege user to persist malicious script in a profile field that is rendered without escaping into portal pages; the payload executes in the browser of any administrator viewing those pages, enabling administrator account takeover. Exploitation requires a low-privilege account with access to edit the vulnerable profile field - realistic where ERPNext exposes self-service portal/website accounts for customers or suppliers - plus administrator interaction to load the poisoned page, which bounds risk below that of a pre-authentication, zero-interaction XSS. No public exploit code was identified at time of analysis.
Stored cross-site scripting in Frappe ERPNext allows an authenticated low-privilege user to plant a malicious payload in a record field that is rendered into an HTML attribute context without escaping; the payload executes in the browser of any higher-privileged user who opens the affected page, which can lead to account takeover of that privileged session. Affected deployments are ERPNext before 15.119.0 and 16.x before 16.31.0, with patched releases 15.119.0 and 16.31.0 addressing the flaw. Exploitation requires an authenticated low-privilege account and depends on a victim with greater privileges passively opening the page, which is the primary limiting factor; there is no CISA KEV listing and no public exploit code identified at time of analysis, and no EPSS score was provided, so this should be treated as a credible, actionable priority rather than an emergency.
Stored cross-site scripting in Frappe ERPNext allows any authenticated user with permission to create records to plant a crafted record name that is later rendered unescaped in a report view, executing arbitrary script in the browser session of whichever user opens that report. Exploitation is not remote-unauthenticated: it requires valid credentials plus create-record rights (CVSS 4.0 PR:L) and is further gated on a second user subsequently opening the vulnerable report (UI:P), meaning the payload lies dormant until then. Impact spans the confidentiality and integrity of the affected session and can be severe when a high-privileged administrator views the report; no public exploit code has been identified at time of analysis and there is no CISA KEV listing, though the issue is rated 8.5 and warrants prompt patching.
Stored cross-site scripting in Frappe ERPNext versions before 15.112.0 and 16.x before 16.23.0 allows an authenticated user who can create the affected record type to embed script in a record name, which is then interpolated unescaped into markup by a tool view and executed in the browser session of any user who later opens that tool. The CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:L/UI:P) and our independent CVSS 3.1 assessment (AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N) agree that exploitation is authentication-gated at low privileges and depends on victim interaction, making this a moderate-priority, insider-reachable issue rather than an unauthenticated emergency; no public exploit code and no confirmed active exploitation were identified at time of analysis. Because the payload persists in stored data, a single malicious record name can affect every user who subsequently visits the vulnerable tool view, so scope is limited to session-level script execution in the victim's context rather than direct server compromise.
Server-side request forgery in Linux-based Progress MarkLogic Server 11.x and 12.x allows an authenticated, low-privileged user to coerce the database server into making HTTP requests toward internal endpoints reachable from its network position, which in cloud-hosted deployments can expose instance metadata services and the cloud credentials they hold. Exploitation requires valid low-privilege credentials (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N) plus a deployment where an internal metadata or credential service is actually reachable, so on-premises or otherwise isolated installations gain no credential-theft payoff and no unauthenticated path exists. Only releases before 12.1.0 (12.x) and before 11.3.7 (11.x) are affected, and no public exploit code or confirmed active exploitation was identified at time of analysis.
Server-side request forgery in Progress MarkLogic Server 12.x (before 12.1.0) and 11.x (before 11.3.7) when hosted on Microsoft Azure lets an authenticated attacker induce the server to issue requests against internal cloud services, potentially exposing Azure credentials or tokens. The flaw carries a CVSS 3.1 base score of 8.5 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N), reflecting a scope change into the cloud control plane and high confidentiality impact; exploitation requires a valid low-privileged MarkLogic account and an Azure deployment, so on-premises and non-Azure instances are outside the described scope. No public exploit code has been identified at time of analysis, and the issue is not listed as confirmed actively exploited (CISA KEV); the highest-value outcome for an attacker is credential or token theft rather than code execution or denial of service.
Local privilege escalation in Edgeless Systems Contrast 1.8.0 through 1.24.0 allows a low-privileged local user on a node to modify the world-writable containerd configuration and overwrite the runc runtime path, leading to high confidentiality, integrity, and availability impact on that node. This is a credible local privilege-escalation issue rather than a remotely wormable threat; the CVSS vector requires local access with low privileges (PR:L), and no public exploit identified at time of analysis.
Privilege escalation in Progress MarkLogic Server affects 11.x releases before 11.3.7 and 12.x releases before 12.1.0, where an attacker who already holds an administrative account can reach authentication material belonging to higher-privileged users and use it to obtain unauthorized administrative access (CWE-269, CVSS 8.4 with scope change). Exploitation is gated by pre-existing admin privileges, so this is primarily an insider or post-compromise escalation risk rather than an externally reachable entry point; no public exploit has been identified at time of analysis and there is no CISA KEV association. No EPSS score was supplied in the available intelligence, so likelihood estimates should be treated as unquantified.
Stored cross-site scripting in EdgeEver through 1.112.0 lets authenticated users who can invoke the application's MCP tools (create_notebook, rename_notebook, create_memo) plant a script payload inside a notebook name or memo title, which is then rendered unsanitized by setMemoDragPreview when the item is dragged, executing attacker script in the app context with access to the desktop bridge's session token. Exploitation requires the victim to perform the drag interaction (UI:A) and the attacker to already hold the privilege needed to call the MCP tools (PR:L), so the payoff is scoped to the desktop bridge session token in affected builds; no public exploit code or confirmed active exploitation was identified at time of analysis. The supplied CVSS 4.0 score of 8.4 overstates real-world urgency for most environments, consistent with the independent assessment's CVSS 3.1 vector (AV:N/AC:L/PR:L/UI:R/S:U/C:H/I:H/A:N), which frames this as a moderate-priority authenticated-attacker issue rather than an unauthenticated emergency.
Navigation allowlist bypass in EdgeEver Desktop versions 1.6.26 through 1.112.0 allows an attacker who can get crafted content rendered in the app (for example a note-template preview) to load an arbitrary origin into the main window and capture the edgeeverDesktop bridge, including the session token, sidecar RPC, and ACP channels. The flaw is an incorrect-authorization issue (CWE-863) in which the will-navigate handler trusts a string-prefix match, so a URL that places the trusted dev-server prefix in the userinfo portion (e.g. https://devserver@attacker.tld) is accepted while the real origin is attacker-controlled. Exploitation requires a low-privileged authenticated foothold to plant the content and user interaction for the victim to open or preview it, and no public exploit code has been identified at time of analysis; the source CVSS 4.0 vector scores 8.4 (High), driven by high confidentiality and integrity impact that can escalate to command execution through ACP adapters.
Stored cross-site scripting in Open WebUI 0.11.1 through 0.11.3 lets any authenticated user who can upload chat attachments plant malicious script inside a DOCX file, which then executes in the browser of a higher-privileged user who previews it - typically an administrator. The injected code can read the administrator's session token from browser local storage and hand it to the attacker, resulting in full takeover of the instance. Exploitation requires a valid low-privileged account plus an admin voluntarily opening the attachment preview, so it is a credible, above-average-priority issue rather than an emergency; no public exploit code has been identified at time of analysis, and no confirmed active exploitation (no CISA KEV listing) has been reported. Fixed in 0.11.4.
Authenticated users with access to the MCP configuration interface in Nginx UI versions prior to 2.4.1 can bypass the directive filtering that the regular HTTP configuration API enforces, injecting restricted Nginx directives such as load_module, js_import, and js_content and then applying them via a configuration reload. Because the attacker can load the NJS module (ngx_http_js_module.so) and import JavaScript they control, they can execute arbitrary code inside the Nginx worker process, giving high impact to confidentiality, integrity, and availability of the managed Nginx instance. Exploitation requires a valid account (per the CVSS vector, PR:L, i.e. authenticated) and depends on the MCP interface being reachable and the Nginx build supporting NJS; no public exploit code has been identified at time of analysis and the issue is not confirmed as actively exploited, but it should be treated as high priority for any deployment below 2.4.1. The vendor-released patch is version 2.4.1, published with security advisory GHSA-5r62-gj48-897x.
Out-of-band XML External Entity (XXE) injection in Eclipse JAXB RI (jaxb-ri) lets an unauthenticated remote attacker read local files when an application unmarshals attacker-influenced XML through the affected JAXB factory helpers (XmlFactory in the core and xsom modules). The pre-patch code relied solely on XMLConstants.FEATURE_SECURE_PROCESSING, so on JAXP/SAX/DocumentBuilder implementations where that single feature does not disable DOCTYPE declarations, external general entities, external parameter entities, or external DTD loading, external entity resolution remains possible and file content can be exfiltrated out-of-band. Exploitation requires both conditions to hold - untrusted XML reaching JAXB unmarshalling, plus a parser stack where secure processing alone is insufficient - which is why the vector carries AC:H and the score is 8.3 (High) on both the supplied CVSS 4.0 vector and the independently assessed CVSS 3.1 (AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:L). No confirmed active exploitation (not in CISA KEV) and no public exploit identified at time of analysis.
Nginx UI versions prior to 2.4.1 fail to enforce the RequireSecureSession step-up check on multiple authenticated mutation endpoints, so an attacker who already holds a valid session token for an OTP- or passkey-enabled account can create or modify automatic backups, DNS credentials and records, certificates and ACME users, Nginx control settings, and site or stream automation without ever supplying a fresh X-Secure-Session-ID. The flaw is a missing-authorization-check class issue (CWE-862) rated CVSS 3.1 base 8.3 (AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:H/A:H): it requires a stolen or otherwise obtained authenticated session (PR:L) but no user interaction, giving high integrity and availability impact. No public exploit code was identified at time of analysis and there is no indication the issue is being actively exploited; the upstream fix is available in release 2.4.1.
Path traversal in IBM Planning Analytics 2.0 and 2.1 allows a remote attacker to read arbitrary files from the host and overwrite certain application files by supplying crafted file paths to the service. The CVSS 3.1 vector (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N, base 8.2) confirms network reachability, low attack complexity, no authentication requirement (PR:N), and no user interaction, with high confidentiality impact and limited integrity impact. No public exploit code and no confirmed active exploitation (CISA KEV) were identified at time of analysis; the vendor has released a fix referenced in the IBM advisory.
Denial of service in Parse Server allows remote attackers to crash the Node.js process by writing an object that contains a file reference without a URL into a class that has Cloud Code save/find/delete triggers or LiveQuery enabled. Unauthenticated exploitation is possible only where the deployment allows public write access or open sign-up; instances enforcing authenticated writes via ACLs or Class-Level Permissions require an account with write rights, so exposure is bounded. The flaw is an uncaught exception (CWE-248) that terminates the server process and can break subsequent requests until restart. No public exploit code was identified at time of analysis, and impact is confined to availability (no confidentiality or integrity loss).
A malicious RDP client can crash a FreeRDP-based RDP server running versions before 3.32.0 by negotiating the NSCodec and requesting an oversized desktop geometry, which forces a buffer-growth allocation failure inside nsc_context_initialize_encode() and leaves dangling plane pointers that are later double-freed during context teardown. The impact is availability-only - no code execution, data exposure, or persistence - and the attack is unauthenticated (CVSS PR:N), but success depends on actually inducing the allocation failure, so the trigger is not perfectly deterministic and the vulnerable server-mode deployment is less common than FreeRDP's client-side use. No public exploit code or confirmed active exploitation has been identified at time of analysis, and the assessed CVSS 3.1 vector (AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H) reflects a moderate-priority crash issue rather than the more alarming 8.2 CVSS 4.0 base score suggests.
Memory corruption in IBM Power Systems host firmware FW1060.00 through FW1060.81 lets an attacker who already holds authenticated, high-privileged service-level access to the Flexible Service Processor (FSP) write crafted configuration data that corrupts the host firmware boot stack during system IPL, with the compromise then propagating to the hypervisor loaded by that firmware and to the managed system. Because the flaw is reachable only from the local FSP interface (CVSS:3.1/AV:L/AC:L/PR:H/UI:N/S:C/C:H/I:H/A:H, base 8.2) and only fires on a (re)boot cycle rather than at runtime, the practical attack path is an insider or compromised FSP service account on the management plane rather than a remote network attacker, and no public exploit identified at time of analysis and no CISA KEV listing is associated with this CVE. Full confidentiality, integrity, and availability impact is scored against the managed system, with a scope change reflecting that the corruption begins in host firmware but affects the separately-managed hypervisor.
Authenticated attackers holding service-level credentials on the Flexible Service Processor (FSP) can read or modify arbitrary regions of Hostboot memory on IBM Power Systems firmware levels FW1060.00 through FW1060.81 by sending a specially crafted message over the service processor mailbox interface, corrupting the host firmware boot stack and the hypervisor it subsequently loads. The flaw is rated CVSS 8.2 (CVSS:3.1/AV:L/AC:L/PR:H/UI:N/S:C/C:H/I:H/A:H), with the scope-change rating reflecting that the FSP compromise crosses a boundary into host firmware and produces high confidentiality, integrity, and availability impact on the managed system. Exploitation is gated behind high-privilege service-processor access rather than network reach to the managed OS, and no public exploit code has been identified at time of analysis; no EPSS score was supplied, so it could not be factored into the risk picture.
Local code execution in IBM Power Systems firmware (FW1060.00 through FW1060.81) lets an attacker who already holds authenticated service-level access to the Flexible Service Processor (FSP) send a crafted service processor mailbox message that overflows host firmware runtime attribute storage, compromising the managed system's memory and hardware interfaces. The CVSS 3.1 vector AV:L/AC:L/PR:H/UI:N/S:C/C:H/I:H/A:H (base 8.2) is internally consistent with that description: the flaw is local, requires high-privilege service-plane credentials, needs no user interaction, and its scope change reflects a jump across the trust boundary from the service processor into the host firmware runtime with high confidentiality, integrity and availability impact. No public exploit code has been identified at time of analysis and there is no CISA KEV entry; IBM published the vendor advisory at https://www.ibm.com/support/pages/node/7290126.
Arbitrary code execution on the host system is possible in IBM Power Systems firmware versions FW1120.00 through FW1120.01, FW1110.00 through FW1110.31, and FW1060.00 through FW1060.81, due to a stack-based buffer overflow (CWE-121) in the host firmware boot process image validation path. An attacker with high-privileged service access to the service processor (PR:H, AV:L) can supply a maliciously crafted code update image to execute arbitrary code on the host system, resulting in full confidentiality, integrity, and availability impact (CVSS 3.1 base score 8.2, High). No public exploit has been identified at time of analysis, and exploitation requires local/service-level access and high privileges, substantially limiting real-world risk.
Arbitrary code execution in the host firmware runtime of IBM Power Systems is possible via the service processor mailbox interface, affecting firmware levels FW1060.00 through FW1060.81. An attacker with authenticated service-level access to the Flexible Service Processor (FSP) can send a crafted mailbox message that overflows host firmware runtime attribute storage (CWE-121), gaining full access to the managed system's memory and hardware interfaces. This is a high-severity local/management-plane issue (CVSS 8.2) requiring high privileges; no public exploit has been identified at time of analysis, and while it is not a mass-exploitation priority, isolation of the FSP network and limiting service accounts are recommended.
NetEase Cloud Music for Windows versions 3.1.38 and 3.1.39.205426 accept forged API responses because /eapi/ and /xeapi/ bodies are decrypted with a fixed AES-128-ECB key and no integrity credential. A network-positioned, unauthenticated attacker (MITM, rogue proxy, or spoofed DNS/ARP) can construct valid ciphertext with PKCS#7 padding, achieving high integrity impact (VI:H) on the client's data flow. The flaw alone only enables forged-response acceptance and requires chaining-such as to the os.navigateExternal bridge-for meaningful exploitation; no public exploit code or active exploitation has been identified at time of analysis.
Path traversal in ConvertX (all releases prior to 0.18.0) lets an authenticated user write converted output files to arbitrary paths inside the container by supplying traversal sequences in the convert_to parameter of the /convert endpoint. The CVSS 3.1 vector AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H reflects that exploitation requires a valid account by default, though instances deliberately configured for unauthenticated access lower that bar to anyone who can reach the service; the payoff is integrity and availability impact - overwriting files or causing denial of service - rather than data disclosure. A vendor fix is available in 0.18.0 (upstream PR 532, commit 0965928), and no public exploit code or confirmed active exploitation was identified at time of analysis, keeping this a credible above-average-priority issue rather than a top-tier emergency.
Credential exposure in Telegram-Stremio through 5.1.0 lets any party with read access to the addon's MongoDB database, its backups, or its application logs recover the operator's stored Telegram session string by running it through the project's own keyless decode_string routine, granting full control of the linked Telegram account even when two-step verification is enabled. Only deployments that have completed a Telegram login (so a session string is actually persisted) are affected, and exploitation is not remotely triggerable against a default running service: the CVSS 4.0 vector carries AT:P, reflecting the prerequisite that the attacker already holds read access to a backend data store holding the string. The published CVSS 4.0 score is 8.2 (High) with confidentiality as the sole impacted metric, no public exploit code has been identified at time of analysis, and the independent assessment rates it CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N.
Out-of-bounds reads in the IBM i File Server allow a remote authenticated attacker to crash or disrupt the service while also leaking portions of adjacent memory, affecting IBM i releases 7.3, 7.4, 7.5 and 7.6. Exploitation is not unauthenticated: the CVSS vector (AV:N/AC:L/PR:L/UI:N) confirms that valid low-privilege credentials and network reachability to the File Server are both required, with high confidentiality and availability impact but no integrity impact, and no user interaction. No public exploit code has been identified at time of analysis, EPSS is low at 0.41% (33rd percentile), CISA SSVC rates exploitation as none and the issue as not automatable, and a vendor patch is available, so this is a moderate-priority patching item rather than an emergency.
Authenticated remote code execution in IBM Planning Analytics 2.0 and 2.1 stems from a stack-based buffer overflow (CWE-121) that allows a remote attacker holding valid low-privilege credentials to corrupt stack memory and execute arbitrary code. IBM assigns a CVSS 3.1 base score of 8.1 (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N), and the independent assessment raises the availability impact to High, indicating a successful exploit may also crash or destabilize the affected service. No public exploit code has been identified at time of analysis and the vulnerability is not listed in CISA KEV; vendor patches are available via IBM support node 7291604.
Session token disclosure in IBM TRIRIGA Application Platform 5.0.0 through 5.0.8 allows a remote authenticated attacker to read session tokens that are rendered directly into HTML response bodies, which can enable session hijacking of whichever account owns the exposed token. The assessed vector is CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N, so exploitation requires a valid low-privileged account and the confirmed impact is loss of confidentiality, not direct integrity or availability damage; the practical severity depends on whether tokens belonging to higher-privileged users are exposed rather than only the attacker's own session, in which case impact is minimal. No public exploit code or confirmed active exploitation has been identified at time of analysis, and a vendor patch is available per the IBM advisory.
MarkLogic Server 11.x before 11.3.7 and 12.x before 12.1.0 can write security-sensitive values - information that enables access to protected resources - into diagnostic logs when the server is running in specific, non-default configurations. Anyone who can already read those logs (filesystem access, a log-aggregation/SIEM pipeline, backup or archive copies, or an account permitted to view logs) can harvest that data and use it to obtain unauthorized access, so the exposure is genuine but gated on two preconditions: a particular logging configuration must be present, and the attacker must already hold low-privileged log read access. No public exploit code or confirmed active exploitation (no CISA KEV listing) was identified at time of analysis, and the independent assessment rates this as a confidentiality-only, local-context issue (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N), making it a moderate-priority hardening problem rather than an emergency despite the 8.1 vendor score.
Authorization bypass in Tuleap Enterprise Edition 17.0 through 17.5 allows an authenticated attacker who already holds tracker administration rights on one tracker to update or delete fields belonging to other trackers on the same instance. The flaw is a CWE-639 user-controlled key authorization bypass rated CVSS 8.1 with high integrity and availability impact, but it requires a low-privilege tracker-admin role (PR:L) and affects only the stated versions. No public exploit identified at time of analysis, and no confirmed active exploitation, though practical risk is highest in large multi-team deployments where many trackers share one instance.