Unauthenticated denial of service in the TP-Link Tapo C325WB v2 camera allows a crafted pair of RTSP-over-HTTP tunneling requests sent to the streaming service on TCP port 554 to corrupt memory (stack-based buffer overflow, CWE-121) and crash the streaming daemon, disrupting live video and related streaming functions until the daemon recovers or restarts. Only cameras with the optional Camera Account feature enabled are exposed, and the attacker must be positioned on the same adjacent network segment (Wi-Fi or LAN) as the camera, since the assessed vector is CVSS:3.1/AV:A/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H; no authentication or user interaction is required. Impact is availability-only and temporary, no public exploit identified at time of analysis, and a vendor patch is available per the TP-Link firmware release notes and advisory FAQ 5333.
Account takeover of Eclipse Ditto Explorer users is possible via a crafted link that injects an attacker-controlled OIDC authority or API URI when the autoSso feature is enabled (versions 3.6.0 through 3.9.7). The victim's OAuth authorization code and PKCE verifier are redirected to the attacker's token endpoint, enabling redemption for access and refresh tokens; alternatively, bearer tokens or Basic credentials are leaked to an attacker-controlled api_uri. Exploitation requires unauthenticated remote attackers to convince a victim to click the link and authenticate, but the configuration persists, causing repeated token theft on subsequent visits; no public exploit code has been identified at time of analysis, and the independent assessment deems this a credible but conditional account-takeover flaw rather than a mass-exploitation emergency.
Authenticated non-admin users of Apache DolphinScheduler 3.2.0 up to (but not including) 3.4.3 can retrieve the Kubernetes kubeconfig that administrators stored for cluster integration, because the API that serves that data performs no authorization check (CWE-862). With the disclosed kubeconfig an attacker can authenticate directly to the Kubernetes API from outside DolphinScheduler, and the real-world severity is bounded by whatever privileges the embedded service account holds - cluster-admin-grade credentials expose Kubernetes Secrets and allow pod creation and persistence, while narrowly scoped credentials sharply limit what is achievable. No public exploit code or confirmed active exploitation (CISA KEV) was identified at time of analysis, and the issue is unreachable on deployments that never configured Kubernetes cluster integration or where the requester cannot obtain even a low-privileged account.
Read-only management accounts on Brocade Fabric OS switches can escalate to restricted administrative operations by submitting management-protocol requests whose non-standard action opcodes the RBAC validation engine mis-categorizes during its permission check; the flaw affects Fabric OS releases before 10.0.1 and is scored CVSS 4.0 base 7.1 (CWE-269, improved privilege management failure). Exploitation is gated: an attacker must already hold a valid authenticated read-only management account (PR:L) and be positioned on an adjacent/management network (AV:A) to reach the affected management plane, with no user interaction required (UI:N). No public exploit code and no confirmed active exploitation were identified at time of analysis, so practical risk is driven by the attacker's existing credentials and management-network access rather than by remote unauthenticated reachability.
Broken authorization in Apache DolphinScheduler before 3.4.3 allows any authenticated user to retrieve connection details - including stored data source passwords - for data sources they are not authorized to access via the /unauth-datasource and /authed-datasource endpoints. Exploitation is network-reachable (AV:N), low complexity (AC:L), requires no user interaction (UI:N) and only a valid low-privileged account (PR:L), yielding high confidentiality impact (C:H) with no integrity or availability impact per our assessment; NVD/vendor scoring differs slightly by also crediting low integrity impact (I:L). Because the leaked credentials can be reused against the underlying databases, the practical consequence extends beyond information disclosure to potential downstream database compromise; no public exploit code or confirmed active exploitation was identified at time of analysis, and the vendor-released fix is version 3.4.3.
Brocade Fabric OS Management Server before version 10.0.1 can be manipulated by a compromised or rogue Fibre Channel switch that is already trusted within the fabric, which uses crafted inband vendor-unique Common Transport (CT) management requests to bypass administrative authentication. This lets the unauthorized peer perform sensitive administrative actions on the target device, including resetting administrative passwords, rebooting the system, and initiating firmware downloads. The attack is confined to the Fibre Channel fabric and cannot be launched from the management IP network or the internet; it requires a pre-compromised fabric peer (assessed AV:A/AC:H with an AT:P precondition in the vendor vector), and no public exploit code has been identified at time of analysis.
Command-line argument injection in SumatraPDF 3.5.2 and earlier lets an attacker-controlled SyncTeX source filename escape the quoting of an external editor's command template and inject arbitrary flags when a victim triggers inverse search. The attack is strictly conditional: it requires a PDF paired with a crafted sibling .synctex.gz file, a configured or auto-detected inverse-search editor, and a deliberate user action (double-clicking to jump to source) at a coordinate that maps to the malicious node. Per the advisory-supported conditions, code execution is not guaranteed by SumatraPDF itself and depends on the target editor honoring the injected flags (for example, installing a malicious extension), so this is a moderate, conditional risk rather than a drop-everything priority; no public exploit has been identified at time of analysis, and no fixed release is available as of this review, though an upstream fix commit exists.
Argument injection in the AI translation feature of SumatraPDF 3.6.1 and earlier allows text placed in a document selection or clipboard to be interpreted as command-line flags when an agentic CLI backend (Grok, Claude Code, or Codex) is configured as the translation provider. Because the vulnerable code builds the child process command line with quote-only escaping, attacker-controlled text can smuggle in flags that alter the model, working directory, approval mode, or sandbox settings used by the launched CLI. Exploitation is local and requires the user to actively invoke the translate/AI-chat action on the crafted text, and only affects installations where one of those agentic CLI backends is installed and selected; no public exploit code was identified at time of analysis, and the issue is not in CISA KEV.
Denial of service in MongoDB libmongocrypt occurs when KMS endpoint parsing triggers an integer underflow, causing an allocation failure that terminates the application process. This can be triggered by an authenticated user with write access to the key vault collection modifying a key document, or when an application passes a KMS endpoint containing a colon after its path or query during key creation. No public exploit code has been identified at time of analysis, and the vulnerability does not allow code execution, data disclosure, or data tampering-only a crash.
Reflected cross-site scripting in the aBlocks plugin for WordPress (versions through 2.16.0) lets an unauthenticated remote attacker run arbitrary script in a victim's browser by enticing them to open a crafted link, as the malicious input is echoed back in the generated page rather than stored (CVSS 7.1, CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:L). Exploitation hinges on victim interaction, and the highest-impact scenario is luring a logged-in administrator, whose hijacked browser session can be used to perform privileged actions against the site. No public exploit code has been identified at time of analysis and the CVE is not listed in CISA KEV; no EPSS score or known exploitation activity was supplied in the available intelligence.
Cluster-wide denial of service in OpenStack Mistral (all versions through 23.0.0) lets any authenticated holder of a valid Mistral token - regardless of role or tenant - read and alter the service's global maintenance state through the /v2/maintenance API, because the controller clears the request context and calls the maintenance service without any policy enforcement. Setting the state to PAUSED halts processing of new workflow and execution objects for every tenant project until an operator intervenes, giving a lowest-privileged tenant member a persistent, cross-tenant availability impact (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H, 7.1 High; NVD's CVSS 4.0 vector scores identically at 7.1 High). Exploit path is reachable against default configurations with no feature flag or non-default setting required, but it is not anonymously exploitable - a valid Mistral API token is the single prerequisite. No confirmed active exploitation (not in CISA KEV) and no public exploit code has been identified at time of analysis; patch availability is not confirmed in the provided data.
Cross-site request forgery in the CISA Malcolm network traffic analysis platform lets a remote attacker who lures the operator into loading attacker-controlled web content force that operator's browser to fire unauthenticated management commands at the kiosk's POST /script_call/<script> endpoint. Successful abuse yields high integrity and availability impact (I:H/A:H) with no confidentiality impact (C:N), because the reachable commands are destructive rather than data-stealing - most notably control.py --wipe, which permanently deletes all captured network traffic and forensic logs, and control.py --stop, which blinds security monitoring. The endpoint ships with zero authentication and wildcard CORS (CORS(app)) in its default state, so no server-side misconfiguration is required; the only real barrier is inducing the specific operator to render a malicious page (UI:R). No public exploit code and no confirmed active exploitation were identified at time of analysis.
Server-side request forgery (SSRF) in the Malcolm file-upload component allows authenticated low-privilege attackers to force the server to download arbitrary URLs via the unmodified upstream FilePond PHP fetch API route, then read back the fetched content through the transfer directory. The flaw, identified as CWE-918, affects Malcolm deployments using the file-upload feature and can expose internal network services or cloud metadata due to curl's FOLLOWLOCATION behavior. No public exploit identified at time of analysis; exploitation requires access to the upload endpoint (PR:L), and impact is limited to confidentiality readback plus minor availability (storage consumption).
Authenticated attackers with a Zimbra Collaboration Suite account can exploit an authorization flaw in the GrantRightsRequest SOAP operation to grant the loginAs right to another local account, creating persistent mailbox access and mail-sending authority that survives password changes and session expiry. The vulnerability requires only low privileges (PR:L) and no user interaction, affecting all versions of Zimbra Collaboration Suite per the CPE. No public exploit or active exploitation has been identified at time of analysis, but the moderate CVSS score of 7.1 reflects the high confidentiality impact and the stealthy, persistent nature of the granted access.
An authenticated low-privileged member of a pH7Builder (pH7 Social Dating CMS) site running a version before 18.5.1 can upgrade themselves to the most expensive paid membership tier without paying the corresponding price. The payment controller trusts client-supplied plan and amount parameters - specifically item_number, cart_order_id, and the PayPal 'custom' field - so an attacker can pay a token amount (or submit an uncompleted PayPal IPN notification) and still be granted the highest tier and all of its paid features. The flaw requires a valid member account (CVSS 4.0 PR:L) and an instance that actually sells paid tiers through an enabled gateway, notably PayPal; no public exploit code or active exploitation has been identified at time of analysis, and the vendor has released a fix in 18.5.1.
Out-of-bounds heap reads in the ZRLE decoder of GlavSoft TightVNC Viewer for Windows (versions before 2.8.88) let a malicious or compromised VNC server crash the viewer and leak small amounts of adjacent heap memory into the rendered framebuffer. The victim must initiate the VNC session to the attacker-controlled or man-in-the-middle'd server and the connection must negotiate ZRLE encoding; the server then sends palette-based tiles whose colour indices exceed the declared palette size, causing readPaletteRleTile() and readPackedPaletteTile() to index past the end of the palette allocation. The assessed vector is CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:N/A:H (7.1), a medium-to-high client-side flaw whose score is driven almost entirely by availability impact (viewer termination) plus a limited confidentiality leak (heap bytes rendered as garbled colours), with no code execution achieved; no public exploit code has been identified at time of analysis, and vendor patches are available in 2.8.88.
Cross-site request forgery in the WPMU DEV Forminator WordPress plugin, affecting all versions through 1.57.3, lets an unauthenticated attacker craft a malicious page that, when visited by a logged-in site administrator, silently submits state-changing requests to the vulnerable site on that administrator's behalf. The CVSS 3.1 score is 7.1 (AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:H/A:N), reflecting high integrity impact offset by the required victim interaction and no availability impact; no public exploit code and no CISA KEV entry were identified at time of analysis, and no EPSS score was supplied. Risk is bounded by the interaction requirement: the attack fails if the targeted privileged user is not currently authenticated in a browser session while visiting the attacker-controlled page.
Stored cross-site scripting can be planted on WordPress sites running the WPComplete plugin through version 2.9.5.6, because a WPComplete write action is performed without adequate CSRF nonce validation (CWE-352) and the stored value is later rendered without sufficient output encoding. Per the CVSS vector (AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:L) the attacker needs no credentials of their own, but exploitation requires tricking a logged-in, privileged WordPress user into visiting an attacker-controlled page or link while their session is active, which is the primary limiting factor. No public exploit code or confirmed active exploitation was identified at time of analysis, and the net effect is a moderate-priority risk: injected script executes in the site context (scope changed) and can affect other users, but impact to confidentiality, integrity, and availability is each rated Low.
Command injection in the Brocade Fabric OS firmware management daemon lets a management user with high privileges execute arbitrary shell commands with the daemon's elevated privileges on switches running Fabric OS before 9.2.2d or 10.0.0 through 10.0.0a1. Because the CVSS vector is local (AV:L), requires high privileges (PR:H) and an operator-driven upgrade workflow (UI:P/UI:R), this is a privilege-boundary flaw for an already-authorized administrator rather than unauthenticated remote takeover, though successful exploitation yields full confidentiality, integrity and availability impact confined to the affected Fabric OS instance. No public exploit code and no confirmed active exploitation (CISA KEV) were identified at time of analysis.
Authenticated low-privilege local users on Brocade Fabric OS switches can abuse the diagnostic kernel module subsystem to invoke privileged hardware tests, force system error conditions, perform hardware blade resets, and disrupt storage fabric operations, because those interfaces ship with improper file permissions and lack authorization checks (CWE-732). Only devices running FOS builds before 9.2.2d, or in the 10.0.0 through 10.0.0a1 range, are affected, and exploitation requires an existing account with shell/CLI access on the switch rather than unauthenticated network reachability. No public exploit code or confirmed active exploitation has been identified at time of analysis (no CISA KEV entry), and the CVSS:3.1 assessment (AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H) places this at moderate risk: the impact is confined to integrity and availability, with no confidentiality or data-disclosure component.
Local privilege escalation in Brocade Active Support Connectivity Gateway (ASCG) before 3.5.0 stems from a well-known Brocade default password embedded in a script that is distributed to every customer, allowing any local authenticated user with read access to the installation path to recover the credential and elevate privileges on Open Virtual Appliance (OVA) deployments where default configuration settings remain in place. The flaw is rated 7.0 (High) under CVSS 4.0 with a local vector (AV:L, PR:L, UI:P), meaning an attacker must already hold a low-privilege local account and passive user interaction is required, so it is not remotely reachable. No public exploit code and no confirmed active exploitation were identified at time of analysis, and the realistic risk is concentrated on OVA deployments that have not rotated the default credential.
Denial of service in MIT Kerberos KDCs (krb5 through 1.22.2) lets a malicious or compromised cross-realm trusted KDC crash krb5kdc by sending an S4U2Proxy request whose PAC carries a malformed client name, interrupting authentication for the affected realm until the daemon restarts. The root cause is a NULL pointer dereference (CWE-476) in get_pac_princ_with_realm(), which reports success while leaving the client principal NULL for malformed names. Exploitation is gated by a significant prerequisite - the victim realm must have a cross-realm trust and the attacker must control or have compromised the KDC in that trusted foreign realm (PR:H in both the CVSS 4.0 and independently assessed CVSS 3.1 vectors) - so single-realm deployments without trusts are not exposed; no public exploit was identified at time of analysis and the issue is not listed in CISA KEV.
Missing rate limiting in Jivejdon's unauthenticated /account/smsVRAction endpoint allows remote attackers to send unlimited SMS messages to arbitrary phone numbers, leading to harassment and depletion of the operator's Tencent Cloud SMS balance. Affected commits range from e0306088 through ee67a65e, and exploitation requires the instance to have Tencent Cloud SMS integration configured and funded; attackers must first load newAccount.jsp to set session attributes before repeatedly calling the endpoint. No public exploit code or active exploitation has been identified, and this is assessed as a medium-severity resource and financial abuse issue rather than a high-priority emergency.
Frappe Cloud and ERPNext fail to properly validate the email field in the self-service registration endpoint /api/method/press.api.account.signup, allowing unauthenticated remote attackers to supply multiple email addresses and have the OTP code delivered to all of them as long as one address is valid. This can lead to OTP interception, impersonation during registration, creation of accounts using other people's email addresses without mailbox access, and indirect confirmation of already registered emails. The issue is rated moderate (CVSS 4.0 6.9, CVSS 3.1 6.9 with low confidentiality and integrity impact); no public exploit identified at time of analysis, and impact is limited to deployments that expose open self-service signup.
SQL injection in Jivejdon through 5.0 allows an authenticated user with the Admin role to read arbitrary database contents, including other accounts' password hashes, by submitting a crafted username to the /admin/user/userListAction endpoint (AccountDaoSql.getAccountByNameLike()). The CVSS 4.0 vector (PR:H) confirms this is not reachable by anonymous or low-privilege users, so the practical risk is insider abuse or post-compromise credential harvesting rather than broad exploitation. No public exploit code or CISA KEV entry is identified in the provided data, and the assessed impact is confidentiality-only (database read; no integrity or availability effect).
Authentication bypass in Jivejdon through 5.0 allows remote attackers to log in as any user whose account was created through the Sina Weibo OAuth flow, because OAuthAccountServiceImp.transferSina() and the SinaUserCallBackAction callback assign those accounts a password made up of the first four digits of the user's public Weibo ID. Since the Weibo ID is publicly visible, an unauthenticated attacker who knows or enumerates a victim's ID can submit that four-digit value through the ordinary form login and then read or post content as the victim (limited confidentiality and integrity impact). No public exploit code was identified and the CVE is not listed in CISA KEV, so exploitation is currently a design-flaw/credential-prediction risk rather than a confirmed actively exploited campaign; only accounts provisioned via the Weibo integration are exposed, not accounts created by other means.
Unauthenticated remote attackers can trigger a use-after-free in IBM DataPower Gateway, resulting in a denial of service and potential disclosure of sensitive information held in freed memory. The flaw affects the 10.5.0 branch (10.5.0.0 through 10.5.0.22), the 10.6.0 branch (10.6.0.0 through 10.6.0.10), the 10.6CD/10.6.1 branch (10.6.1 through 10.6.6), and the 11.0.0 branch (11.0.0.0 through 11.0.0.2); the CVSS vector (AV:N/AC:L/PR:N/UI:N, C:H/I:N/A:H) rates it 9.1, but exploitation requires the gateway endpoint to be network-reachable and, as a memory-safety bug, reliable triggering typically depends on request sequencing and allocator/timing state. There is no public exploit identified at time of analysis, EPSS is low at 0.24% (14th percentile), and CISA's SSVC framework records exploitation as none and automatable as no - signals are therefore mixed between a high headline severity score and a low observed exploitation likelihood. A vendor patch is available per IBM's advisory.
Out-of-bounds memory handling in IBM DataPower Gateway (10.5.0.0 through 10.5.0.22, 10.6.1 through 10.6.6, 10.6.0.0 through 10.6.0.10, and 11.0.0.0 through 11.0.0.2) allows network attackers to trigger a buffer overflow that can disclose sensitive process memory or crash the appliance. Per the assessed vector (AV:N/AC:L/PR:N/UI:N), exploitation is unauthenticated and requires no user interaction, with high confidentiality and availability impact but no integrity impact - consistent with the CWE-125 out-of-bounds read classification rather than a write-based code-execution primitive. No public exploit code was identified at time of analysis, CISA SSVC records exploitation as 'none' and automation as 'no', and EPSS is low at 0.22% (12th percentile); note that the vendor's own 9.8 rating assumes full C/I/A impact, which this assessment does not corroborate. Patch available per vendor advisory (IBM support page node/7289775).
Improper neutralization of CRLF sequences in the Federated Authentication (FA) configuration of Brocade Fabric OS before 10.0.1 lets an authenticated administrator - or anyone able to supply a crafted FA configuration file that an operator imports - inject arbitrary web server directives through the Identity Provider (IdP) issuer parameter, which can prevent the web management daemon from starting (denial of service) or alter web server security controls. Exploitation is heavily constrained: the target must run a vulnerable Fabric OS release with FA actually in use, the attacker needs administrative privilege (CVSS PR:H) and adjacent-network access to the management interface (AV:A), with no user interaction otherwise required, and no public exploit code has been identified at time of analysis. This is a medium-severity, operator-dependent flaw rather than a top-priority emergency, and it is not confirmed actively exploited (CISA KEV).
An authorization-logic bypass in the Brocade Fabric OS web management framework (all versions before 10.0.1) allows an authenticated, low-privileged user to circumvent inner Role-Based Access Control checks and lower the authorization mode of their active session, gaining access to configuration settings reserved for administrative roles. Exploitation is not trivial: it requires a valid low-privileged account, adjacent-network (management-plane) access, a vulnerable Fabric OS build, and the additional environmental conditions the vendor cites as the trigger; the weakness affects integrity only and does not by itself yield code execution or data disclosure. This is a MEDIUM-severity issue (vendor CVSS 4.0 base 6.9; independent CVSS 3.1 assessment AV:A/AC:H/PR:L/UI:N/S:U/C:N/I:H/A:N) and no public exploit code or confirmed active exploitation was identified at time of analysis.
An authenticated path traversal (CWE-22) in the Zone Alias Import flow of Brocade SANnav before 3.0.0a allows an attacker who already holds valid SANnav credentials to write uploaded content outside the intended upload directory, corrupting or planting files on the management host. The assessed vector (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:N) confines impact to integrity only, with no confidentiality or availability effect and no user interaction required; per the vendor's CVSS 4.0 vector the attack surface is the management interface reached from an adjacent network segment. There is no evidence of active exploitation (not in CISA KEV) and no public exploit code identified at time of analysis, so this is a moderate-risk, authenticated file-write issue rather than a top-priority emergency.
Denial of service in Brocade Fabric OS before 10.0.1: an authenticated user holding administrative privileges on the switch can crash the diagnostic execution utility by passing a diagnostic command string containing an excessive number of tokenized arguments, which overflows a fixed-capacity argument array on the stack and terminates the process. The issue is reachable only from an adjacent network position such as the management LAN (AV:A) and requires pre-existing admin credentials (PR:H, no user interaction), so it is a medium-severity availability problem for already-privileged operators rather than a remote, unauthenticated compromise path; confidentiality and integrity of the device are unaffected, with the availability impact rated High. No CISA KEV entry or publicly available exploit code was identified, and no public exploit identified at time of analysis; EPSS data was not provided in the available intelligence.
Stack buffer overflow in Brocade Fabric OS WebTools administrative interface allows an authenticated user with permissions to perform configuration downloads using remote server profiles to crash the weblinker daemon, resulting in a privileged denial of service. Affected versions are Fabric OS before 9.2.2d and 10.0.0 through 10.0.0a1. Exploitation requires adjacent network access (AV:A) and high privileges (PR:H); no public exploit code has been identified at time of analysis.
Denial of service in Dislocker through 0.7.3 is triggered when the tool parses a crafted BitLocker volume image whose VMK or FVEK datum declares a datum_size smaller than the 36-byte AES-CCM header it must contain, causing an integer underflow that yields a near-4GB payload length and sends hexdump() into an out-of-bounds heap read that crashes the process. Exploitation is local and requires user interaction: a victim must actively point dislocker at the attacker-supplied image, which matches the assessed vector CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H, unauthenticated but victim-driven. Impact is availability only - there is no memory corruption of confidentiality or integrity consequence described, no public exploit identified at time of analysis, and the defect is not listed as confirmed actively exploited (CISA KEV).
Denial of service and heap memory disclosure in Dislocker 0.7.3 and earlier: a crafted BitLocker volume image whose metadata announces inflated dataset or datum sizes pushes the parsing routines get_dataset() and get_next_datum() past the end of the allocated metadata buffer, crashing the tool or leaking adjacent heap memory through the volume-parsing output. Any default build of Dislocker at or below 0.7.3 that opens the malicious image is affected, which is of practical concern mainly to forensic, incident-response, and disk-recovery workflows that routinely process untrusted BitLocker volumes. Exploitation is local and requires the victim to actually open or mount the crafted image (UI:P), and no public exploit code has been identified at time of analysis; an upstream fix commit is available. This is a genuinely Medium-severity, lower-priority issue rather than an urgent one, and it is not noted as confirmed actively exploited (CISA KEV).
Brocade Fabric OS before 10.0.1 contains a stack-based buffer overflow in its security library when parsing uploaded X.509 PEM certificates. An authenticated administrator can trigger the overflow by importing a certificate with an oversized Authority Key Identifier (AKI) extension, crashing the management daemon and causing a denial of service. No public exploit code or active exploitation has been identified at time of analysis; the CVSS 3.1 score is 6.9 (medium), reflecting the need for adjacent-network access and admin credentials.
Unauthenticated path traversal in the Ollama `/api/pull` endpoint lets a remote attacker write an arbitrary file outside the model store by submitting a crafted path-traversal sequence in the layer digest field, because the `digestToPath` function does not adequately validate digests before constructing a filesystem path. When the Ollama process can write to `/usr/lib/ollama` (described as the default in most Ollama Docker images) and the server or container is subsequently restarted, the planted binary is loaded and executed, yielding remote code execution as root; the same file-write primitive is available without authentication to anyone who can reach the API. The issue is fixed in Ollama 0.35.0, no public exploit code has been identified at time of analysis, CISA's SSVC assessment records exploitation as 'none', and no EPSS score was supplied in the source data. Real-world risk is high but conditional: exploitation of the full RCE chain depends on write access to `/usr/lib/ollama` and a restart, so deployments not using the default Docker image, running Ollama as a non-privileged user without write access to that directory, or not exposing the API to untrusted networks face substantially reduced exposure.
Integer overflow in containerd overlaybd's LSMT index loader (do_load_index) lets a crafted, untrusted overlaybd blob pulled from a registry corrupt heap state and crash the host-side runtime, degrading availability for all overlaybd-backed containers on the node. Affected are overlaybd builds through 1.0.18 that are actually used as the container snapshotter/backend; exploitation is unauthenticated (CVSS PR:N) but requires high attack complexity (AC:H) - the victim must load an attacker-controlled blob in a deployment running multiple overlaybd-backed containers, which is why Scope is Changed. Impact is availability-only (C:N/I:N/A:H, CVSS 6.8); there is no public exploit identified at time of analysis and no CISA KEV confirmation, and organizations pulling only from trusted, signed sources are not realistically exposed.
Plaintext HTTP responses can plant, overwrite, or delete Secure cookies held in AsyncHttpClient's ThreadSafeCookieStore, because the store never considers the scheme a Set-Cookie arrived on, so the attacker's value is subsequently sent back to the HTTPS site inside TLS. Affected are the Maven artifact org.asynchttpclient:async-http-client 3.0.0 through 3.0.13 and 2.1.0 through 2.16.1; exploitation is remote and unauthenticated (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:N/I:H/A:N, 6.8, CWE-384) but conditional - the application must have the cookie store enabled and share one store across plaintext and HTTPS origins of the same site, and the HTTPS application must rely on the affected cookie for session or CSRF state. Concretely this yields session fixation into the HTTPS session, a clobbered CSRF token, or the removal of a cookie the site depends on; there is no confirmed active exploitation (not in CISA KEV) and no public exploit identified at time of analysis, and the severity is tempered by the requirement for a plaintext foothold on the same site (a same-site plaintext host setting a Domain cookie, or a network man-in-the-middle).
Server-side request forgery in Indico (all versions prior to 3.3.13) allows an authenticated event organizer to make the server fetch a prohibited local or internal URL by embedding a backslash in a crafted URL, and then read the target's response through affected Indico features. The vulnerability is a CWE-918 SSRF stemming from an incomplete fix for CVE-2026-25738; the trigger (demonstrated as 'http://127.0.0.1\@1.1.1.1') defeats Indico's local-target filter, which now rejects any URL containing a backslash. Exploitation requires the trusted event-organizer role (CVSS 3.1 PR:H, score 6.8, scope changed), there is no confirmed active exploitation (not in CISA KEV), and no public exploit code was identified at time of analysis.
Server-side request forgery in Pydantic AI's URL-fetching guards lets an application that has explicitly opted a URL into local-network access reach IPv6 cloud metadata endpoints such as fd00:ec2::254 and read short-lived cloud IAM credentials, because appending an RFC 4007 zone identifier (for example fd00:ec2::254%251) makes the address unequal to its blocklist entry while the kernel silently ignores the zone for a non-link-local destination and routes the request anyway. Affected releases are pydantic-ai and pydantic-ai-slim at 1.56.0 up to but not including 1.107.6, and 2.0.0b1 up to but not including 2.44.0; exploitation additionally requires that the URL be attacker-influenced or model-chosen, that one of the two opt-ins be enabled (FileUrl with force_download='allow-local', or web_fetch_tool(allow_local_urls=True) - both off by default), and an IPv6-enabled environment with a reachable metadata service. Vendor-released patches are available in versions 1.107.6 and 2.44.0 (upstream fix tracked via pull requests 8401 and 8402); no public exploit code was identified at time of analysis and there is no confirmed active exploitation (CISA KEV).
Stored cross-site scripting (XSS) in the Workflow Template feature of SQLView KRIS lets an actor who already holds administrative access store a script payload in the 'template name' field, which the application renders into an onclick attribute on the main dashboard without server-side sanitisation; the script then executes in the browser of any user who loads that dashboard and triggers the affected element. Per the CVSS vector (AV:N/AC:L/PR:H/UI:R), exploitation requires high administrative privileges and a separate victim's interaction, so the practical exposure is confined to malicious or compromised administrators rather than unauthenticated outsiders, and the independent assessment rates the resulting confidentiality and integrity impact as Low with no availability impact (CVSS 6.8, Medium). There is no KEV entry and no public exploit code identified at time of analysis.
Stored cross-site scripting in the WordPress plugin Image Photo Gallery Final Tiles Grid (versions before 3.6.14) lets an authenticated attacker with contributor-level access or above inject arbitrary JavaScript that runs in the browser session of anyone viewing the affected post. The flaw arises because several gallery configuration values can be overridden through the plugin's gallery shortcode and are printed unsanitised into an inline script block; because no gallery ownership is required, any gallery already present on the site can be referenced, and an administrator previewing a pending contributor submission is the most likely victim. Exploitation requires the victim to load the crafted post (UI:R), there is no public exploit identified at time of analysis, and a vendor-released patch in version 3.6.14 is available.
Malformed Apache configuration data accepted by Brocade Fabric OS during web management service setup or re-initialization can corrupt the stored configuration and prevent the administrative web interface from starting or recovering, producing a persistent denial of service that recurs on each service bring-up. Exploitation requires an attacker who already holds high administrative privileges on the switch and can reach it from an adjacent network segment (assessed CVSS:3.1 AV:A/AC:L/PR:H/UI:N, availability-only impact), and no public exploit code was identified at time of analysis. Only Fabric OS releases before 9.2.2d and the 10.0.0 through 10.0.0a1 branch are affected, with a vendor advisory published by Broadcom.
Information disclosure in AWS CDK's core library (aws-cdk-lib) before 2.267.0 allows a context-dependent actor to cause files from the build host to be published as a deployed asset through improper symbolic-link resolution during asset bundling. Exploitation is local to the build environment and requires the victim to actually run the bundling/synth step (UI:P) on a directory whose contents the attacker can influence, so a developer or CI pipeline must consume untrusted asset sources for the issue to trigger. No public exploit code or confirmed active exploitation has been identified at time of analysis; the primary risk is unintended disclosure of build-host files (for example credentials or configuration) into a deployed stack. Vendor-released patch: 2.267.0, which changes bundling to throw on external symlinks via SymlinkFollowMode.BLOCK_EXTERNAL.
Heap- or stack-based memory corruption in the internal FSPF route-validation and diagnostic routine of Brocade Fabric OS can crash the routing daemon or, worst case, allow code execution in the daemon's context, but only for a local, authenticated attacker who first chains one or more separate vulnerabilities to inject an oversized diagnostic state payload. The flawed code path is explicitly not exposed through standard user interfaces or CLI management commands, so the CVSS 4.0 score of 6.8 overstates standalone risk; our independent assessment rates it medium (CVSS 3.1 AV:L/AC:H/PR:L/UI:N, availability-only impact) and it cannot be triggered in isolation against a default configuration. No public exploit code was identified at time of analysis and there is no evidence of active exploitation; the vendor fix is Fabric OS 10.0.1.
Opening a malicious CHM (compiled HTML Help) file in SumatraPDF 3.6.1 or earlier crashes the application: unsigned file-controlled string offsets from the /#WINDOWS and /#IVB sections are narrowed to signed integers in ChmFile::GetCharZ() without a lower-bound check, so a negative offset drives a read before the /#STRINGS buffer. Exploitation is local and requires the victim to open the crafted file, and the outcome is limited to a deterministic termination (denial of service) with no code execution, data disclosure, or privilege gain; no public exploit code was identified at time of analysis. Cross-referenced with the advisory and the fixing commit, this is a medium-severity availability issue rather than a high-urgency priority.
Opening a crafted TIFF file in SumatraPDF 3.6.1 and earlier can crash the application through a stack-based buffer overflow in MaybeFlipBitmap(), where GDI+ writes an attacker-controlled EXIF PropertyTagOrientation size into a fixed 64-byte stack buffer. The attack is local and requires the victim to open a malicious image (CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P, base 6.8), and successful exploitation corrupts stack control data that the compiler's stack cookie check intercepts in tested builds, terminating the process. Impact is limited to availability (no confidentiality or integrity loss is claimed), with no confirmed active exploitation (not listed in CISA KEV) and no public exploit identified at time of analysis.
Cleartext storage of HTTP proxy credentials in Brocade Fabric OS SupportLink diagnostic support bundles (all versions before 10.0.1) lets anyone who can read a generated bundle recover the proxy account's username and password. The issue is access-gated rather than network-reachable: it only manifests when SupportLink is configured to use an authenticated HTTP proxy, and the attacker must already have read access to the bundle, either as support personnel handling it or as a user able to reach the file share where bundles are retained. Impact is limited to confidentiality of the proxy credentials (CVSS 4.0 score 6.8; assessed CVSS 3.1 AV:L/AC:L/PR:L/UI:N/C:H/I:N/A:N) with no integrity or availability effect, and no public exploit identified at time of analysis.
A local null pointer dereference in SumatraPDF 3.6.1 and earlier lets any process in the same interactive Windows session that runs at an integrity level equal to or higher than SumatraPDF's (the Windows UIPI requirement for sending it inter-process messages) terminate the reader by dispatching a CmdExec command over DDE or WM_COPYDATA while the application sits on its start page with no document tab open. The sole consequence is abrupt process termination and loss of unsaved state; no code execution, data disclosure, or persistence is claimed, and there is no network-reachable vector. No public exploit identified at time of analysis, and no fixed release version was confirmed at review time, although an upstream fix commit and its regression test are publicly visible.