CVE Feed

Improper link resolution before file access ('link following') in SQL Server allows an authorized attacker to elevate privileges over a network.

Concurrent execution using shared resource with improper synchronization ('race condition') in Windows Ancillary Function Driver for WinSock allows an authorized attacker to elevate privileges locally.

Double free in Microsoft Exchange Server allows an unauthorized attacker to execute code over a network.

Improper limitation of a pathname to a restricted directory ('path traversal') in Windows PowerShell allows an unauthorized attacker to bypass a security feature over a network.

Improper verification of cryptographic signature in Windows RDP Client allows an unauthorized attacker to disclose information over a network.

Heap-based buffer overflow in SQL Server allows an unauthorized attacker to execute code over a network.

Use after free in Windows Kernel allows an authorized attacker to elevate privileges locally.

Missing authorization in Entra ID allows an authorized attacker to elevate privileges over a network.

Concurrent execution using shared resource with improper synchronization ('race condition') in DNS Server allows an unauthorized attacker to execute code over a network.

Concurrent execution using shared resource with improper synchronization ('race condition') in Windows DNS allows an unauthorized attacker to deny service over a network.

Heap-based buffer overflow in Visual Studio allows an unauthorized attacker to execute code over a network.

Use after free in Windows Image Acquisition allows an authorized attacker to elevate privileges locally.

Permissive cross-domain policy with untrusted domains in Azure Arc allows an unauthorized attacker to elevate privileges over a network.

Heap-based buffer overflow in Microsoft Windows Media Foundation allows an unauthorized attacker to execute code over a network.

Use after free in Windows Installer allows an authorized attacker to elevate privileges locally.

Out-of-bounds read in Microsoft Windows Media Foundation allows an unauthorized attacker to execute code over a network.

An insufficient input validation vulnerability in the listed NETGEAR RAX series models allows a network-adjacent attacker having network access (such as WiFi credentials) to crash the router's management UI. There is no confidentiality or integrity impact. A crash of the router's management UI does not impact the availability of the router's core services like WiFi network.

A cross site request forgery (CSRF) vulnerability in the listed NETGEAR models allows an attacker who can leverage social engineering techniques on a router administrator to tamper with router configuration and disrupt router operations with active assistance from the router administrator. There is no confidentiality impact due to this vulnerability.

n8n is an open source workflow automation platform. Prior to 2.37.7 and 2.38.1, the OAuth token endpoint bound an authorization code's first access token to the consented resource but did not bind its refresh token. Refreshing checked only that the requested resource was registered, not that it matched the original grant. An OAuth client approved for one workflow could substitute a different workflow URL in the resource parameter and obtain a valid token for an unapproved workflow accessible to the consenting user. This issue is fixed in versions 2.37.7 and 2.38.1.

Exposure of sensitive information to an unauthorized actor in .NET allows an authorized attacker to elevate privileges locally.

Origin validation error in .NET allows an unauthorized attacker to disclose information over a network.

DeepSeek Harness before 0.1.2-alpha.1 contains an authentication bypass vulnerability that grants unauthenticated access to its local HTTP agent-control API by accepting a client-supplied loopback Host header in place of validating the actual TCP connection origin. On the default configuration, a confined tool-executed process can reach the loopback API without any port exposure and use it to escape its own OS sandbox, escalate to unconfined execution, and disable the approval prompt. When the port is externally reachable via tunnel, SSH forward, or reverse proxy, a remote attacker can exploit the same flaw to create sessions, execute arbitrary commands, and exfiltrate stored conversation transcripts without credentials.

An information disclosure vulnerability has been identified in Omada Controller.  An API endpoint intended for Controller initialization remains accessible after completion and may disclose account-related information to unauthenticated remote users.  Successful exploitation may allow an attacker to remote query the affected endpoint that may facilitate user enumeration and subsequent attacks targeting administrative accounts.

Apache Airflow FAB provider versions 3.7.3 through 3.8.0 do not validate the issuer or audience of Azure AD `id_token`s during OAuth login. Deployments are affected only when the FAB auth manager is configured with Azure AD as an OAuth provider. Because the signing keys are fetched from Microsoft's **multi-tenant** JWKS endpoint, an `id_token` minted in *any* Azure tenant — including one the attacker creates — passes signature verification, and the username and role assignments are then read from that attacker-controlled token. Anyone able to register an Azure tenant can therefore authenticate to the Airflow UI with no prior access to the deployment. The fix for **CVE-2026-59243** was incomplete, and this advisory closes the remaining gap: that fix made the provider verify the `id_token` signature, but did not add issuer or audience checks. Operators who already applied the CVE-2026-59243 fix are **still affected and must upgrade again** — 3.7.3 is the release that shipped that fix, so every version containing it falls inside this affected range. Upgrade to apache-airflow-providers-fab `3.8.1` or later.

A security vulnerability has been detected in aircheng-org iWebShop-5 up to 5.15. The impacted element is the function uploadFile of the file controllers/pic.php. Such manipulation of the argument outerSrc/selectPhoto leads to cross site scripting. The attack may be performed from remote. The exploit has been disclosed publicly and may be used. The project was informed of the problem early through an issue report but has not responded yet.

AshLua exposes Ash read actions to Lua scripts run through an eval action. A read call accepts an operation (list, min, max, first, sum, avg) that builds an ad-hoc Ash.Query.Aggregate over a named field and returns its raw value. Ash field policies redact forbidden fields on returned records (replacing them with %Ash.ForbiddenField{}), but that redaction does not apply to aggregate values. A script could therefore read a field the calling actor's field policies forbid by requesting it as an aggregate instead of as a field. This includes fields that are public? true but restricted per-actor by a field policy, such as sensitive PII. The prior hardening only enforced the exposed-field allow-list (field visibility), which is a separate axis from per-actor field-policy authorization. The fix authorizes the aggregated field against the resource's field policies, so aggregating over a field the actor may not see is refused or scoped to the rows where it is visible. This issue affects ash_lua: from 0.1.0 before 0.2.2.

A NULL Pointer Dereference vulnerability [CWE-476] vulnerability in Fortinet FortiOS 7.4 all versions, FortiOS 7.2 all versions, FortiPAM 1.9.0, FortiPAM 1.8 all versions, FortiPAM 1.7 all versions, FortiPAM 1.6 all versions, FortiPAM 1.5 all versions, FortiPAM 1.4 all versions, FortiPAM 1.3 all versions, FortiPAM 1.2 all versions, FortiPAM 1.1 all versions, FortiPAM 1.0 all versions, FortiProxy 7.6.0 through 7.6.6, FortiProxy 7.4 all versions, FortiProxy 7.2 all versions may allow an authenticated attacker to crash the httpsd daemon via crafted HTTP requests.

An improper access control vulnerability in Fortinet FortiManager 7.6.0 through 7.6.4, FortiManager 7.4.0 through 7.4.10, FortiManager 7.2 all versions, FortiManager Cloud 7.6.2 through 7.6.4, FortiManager Cloud 7.4.1 through 7.4.10, FortiManager Cloud 7.2 all versions may allow an administrator to bypass the approval process for workflow sessions via crafted HTTP or HTTPs requests.

A use of uninitialized variable vulnerability in Fortinet FortiAnalyzer 7.6.3 through 7.6.6 may allow attacker to denial of service via <insert attack vector here>

A improper validation of certificate with host mismatch vulnerability in Fortinet FortiOS 7.6.1 through 7.6.6, FortiProxy 7.6.2 through 7.6.6 may allow attacker to information disclosure via <insert attack vector here>

A improper access control vulnerability in Fortinet FortiSandbox 5.0.0 through 5.0.5, FortiSandbox 4.4.0 through 4.4.8, FortiSandbox Cloud 5.0.4 through 5.0.5, FortiSandbox PaaS 5.0.4 through 5.0.5 may allow attacker to access sensitive information via crafted HTTP requests.

A improper access control vulnerability in Fortinet FortiSOAR PaaS 7.6.0 through 7.6.6, FortiSOAR PaaS 7.5.0 through 7.5.3, FortiSOAR PaaS 7.4 all versions, FortiSOAR PaaS 7.3 all versions, FortiSOAR on-premise 7.6.0 through 7.6.6, FortiSOAR on-premise 7.5.0 through 7.5.3, FortiSOAR on-premise 7.4 all versions, FortiSOAR on-premise 7.3 all versions may allow attacker to escalation of privilege via <insert attack vector here>

AshAi exposes Ash read actions to language-model tool calls. The read tool accepts an aggregate result type (min, max, sum, avg) that builds an ad-hoc Ash.Query.Aggregate over a named field and returns its raw value. Ash field policies redact forbidden fields on returned records (replacing them with %Ash.ForbiddenField{}), but that redaction does not apply to aggregate values. A tool caller could therefore read a field the calling actor's field policies forbid by requesting it as an aggregate; min/max in particular return an actual field value. This includes fields that are public? true but restricted per-actor by a field policy, such as sensitive PII. The tool's existing check only required the field to be public, which is a separate axis from per-actor field-policy authorization. The fix authorizes the aggregated field against the resource's field policies, so aggregating over a field the actor may not see is refused or scoped to the rows where it is visible. This issue affects ash_ai: from 0.1.0 before 1.0.3.

A url redirection to untrusted site ('open redirect') vulnerability in Fortinet FortiSIEM 7.5.0 through 7.5.1, FortiSIEM 7.4.1 through 7.4.2 may allow attacker to execute unauthorized code or commands via <insert attack vector here>

A improper neutralization of special elements used in a command ('command injection') vulnerability in Fortinet FortiSandbox 5.2.0, FortiSandbox 5.0.0 through 5.0.6, FortiSandbox 4.4.0 through 4.4.9 may allow attacker to execute unauthorized code or commands via <insert attack vector here>

A unverified ownership vulnerability in Fortinet FortiClientWindows 7.4.0 through 7.4.7, FortiClientWindows 7.2 all versions may allow attacker to improper access control via <insert attack vector here>

A weakness has been identified in aircheng-org iWebShop-5 up to 5.15. The affected element is the function member_list of the file controllers/member.php. This manipulation of the argument Search causes sql injection. The attack is possible to be carried out remotely. The exploit has been made available to the public and could be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.

An integer overflow in the query planning component of MongoDB Server can allow an authenticated user with ordinary database-level read/write privileges to bypass an internal resource limit. Submitting a specially crafted query causes the server to consume memory without bound during query planning, and the resulting exhaustion terminates the server process. This may result in a denial of service affecting all databases served by the affected node.

An uncontrolled resource consumption weakness exists in the request-handling path of the MongoDB sharded-cluster router process. A client that has network access to a router port and has not authenticated can supply connection-monitoring parameters that cause the server to expend CPU resources without any rate limiting, degrading or denying service to legitimate clients. No authentication, elevated privileges, or user interaction is required. Only availability is affected; data confidentiality and integrity are not impacted.

MongoDB Server contains an incorrect authorization vulnerability in the aggregation framework. An authenticated user with minimal privileges can craft a specially formatted aggregation request that causes the server's authorization subsystem to evaluate a different operation than what is actually executed, resulting in unauthorized read access to collection data within the target database.

A security issue in the MongoDB Server aggregation framework allows an authenticated user with limited read privileges to bypass view-level authorization checks and access data from unauthorized collections when Atlas Search features are in use. The issue stems from insufficient validation of an internal command parameter that can be set by external clients, causing a security check to be improperly skipped.

Insufficient validation of storage engine configuration options in MongoDB Server allows an authenticated user with write privileges to supply crafted parameters during collection creation that override internal storage metadata. This results in an out-of-bounds memory write in the server process, causing a denial of service via server crash, with potential for further impact including arbitrary code execution.

A security issue in MongoDB Server's diagnostic reporting interface allows an authenticated user with monitoring privileges to access insufficiently protected credentials from concurrent administrative operations. The same credentials are properly redacted in server log output, but the diagnostic interface omits equivalent redaction. Successful exploitation requires a valid authenticated session with monitoring-level permissions and results in exposure of cleartext credentials that could enable impersonation of other users, including privileged accounts.

A security issue in MongoDB Server's query statistics serialization on the router allows users with monitoring privileges to access unredacted search query text from other users' operations. An improper conditional check in the serialization logic causes the data redaction mechanism to be bypassed when processing search queries through the sharded cluster router. This results in sensitive query literals being stored and made accessible through the query statistics interface.

A security issue in MongoDB Server allows an authenticated user with write privileges to trigger a persistent fatal assertion crash by sending specially crafted retryable write commands. The crash state is durably persisted, causing the server process to repeatedly crash on restart and potentially propagating to additional nodes in a sharded cluster. Manual intervention is required to restore service availability.

Improper handling of case sensitivity in the configuration validation component of MongoDB Server may cause the authorization subsystem to remain in a default disabled state during server startup. An unauthenticated user with network access to a deployment where this condition occurs can perform arbitrary administrative operations, resulting in full impact of data confidentiality, integrity, and availability.

A heap out-of-bounds read security issue exists in the query planning component of MongoDB Server. An authenticated user with database read and write privileges can trigger the security issue through crafted query operations, causing the server to read memory beyond allocated buffer boundaries. The revealed memory contents may be partially observable through diagnostic query statistics output.

A security issue in the MongoDB Server's storage engine integration layer allows an authenticated user with collection creation privileges to cause a persistent denial of service. Insufficient validation of user-supplied storage configuration options permits values that, once persisted to durable metadata, trigger a fatal assertion failure when the metadata is subsequently read by diagnostic operations. The corrupted metadata persists across server restarts and is replicated to other cluster members, requiring manual operator intervention to restore service.

A security issue in MongoDB Server allows an unauthenticated network user to cause a denial of service on a specific type of replica set member. The server contains an assertion in its read concern processing logic that can be reached without authentication, and the assertion's assumptions about internal state do not hold for all member configurations, causing the server process to terminate.

A use-after-free security issue in the cursor management component of MongoDB Server allows an authenticated user to cause a denial of service. Under specific timing conditions during cursor operations, a stale pointer to a freed resource may be retained and subsequently dereferenced during cursor cleanup, leading to a server process crash.