Improper write restrictions in the App Key Value Store (KV Store) collections used by Splunk Secure Gateway let any authenticated Splunk user who does not hold the 'admin' or 'sc_admin' role modify alert and mobile-device recipient records, which downstream alerting and subscription workflows then consume. The flaw affects Splunk Enterprise below 10.4.3, 10.2.7, 10.0.10, and 9.4.15, and Splunk Secure Gateway below 3.10.11, 3.9.25, and 3.8.72, with an assessed CVSS of 6.3 (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L, CWE-732). No public exploit code has been identified at time of analysis and there is no KEV confirmation of active exploitation; risk is concentrated in deployments that actually run Secure Gateway, and impact is bounded to the integrity and confidentiality of Secure Gateway alert/subscription data rather than arbitrary Splunk data or host-level access.
Email spoofing is possible through OVHcloud's shared mail servers, allowing any authenticated OVH email user to forge the sender address to impersonate any OVH-hosted domain that uses the default SPF record include:mx.ovh.com. This vulnerability affects OVHcloud's email service and is assessed as a moderate, bounded risk because exploitation requires a valid account and is mitigated by DMARC with DKIM alignment. The issue stems from an insecure default SPF configuration that authorizes the shared mx.ovh.com infrastructure for all domains using the default record.
Heap-based buffer overflow in GIMP's PCX export plug-in allows a local attacker to corrupt memory when a user opens or creates an image with extremely large width and height and exports it to PCX format. The flaw, tracked as CVE-2026-106065 with a CVSS score of 6.3 (AV:L/AC:H/PR:N/UI:R), stems from 32-bit integer overflow in the width*height size calculation while GEGL operations use the full extent. Affected GIMP packages ship with Red Hat Enterprise Linux 6 through 10; no public exploit code or active exploitation has been identified at time of analysis.
Heap-based buffer overflow in GIMP's 'Hot' color filter plug-in allows a crafted, extremely large image to corrupt heap memory when the filter is invoked, giving an attacker the ability to crash the plug-in and potentially achieve code execution within the running user's session (CWE-119). The flaw is reachable only locally and requires the victim to open an attacker-supplied image with dimensions so large that the width * height allocation arithmetic overflows 32-bit integers, then actively run the Hot filter on it - hence the assessed vector AV:L/AC:H/PR:N/UI:R with high integrity and availability impact but no confidentiality impact. There is no public exploit identified at time of analysis, no CISA KEV listing, and no EPSS score provided, so this is a targeted, social-engineering-dependent client-side issue rather than an automatable, mass-exploitation threat. Red Hat hosts the primary advisory (access.redhat.com/security/cve/CVE-2026-106067) with tracking in Bugzilla 2547461, and GIMP packages across Red Hat Enterprise Linux 6 through 10 are enumerated as affected via CPE.
Exporting an oversized image through GIMP's raw data export plug-in can under-allocate a heap buffer and cause an out-of-bounds write on the heap, with the failure originating in an integer overflow in the g_malloc() size computation (width × height × bytes-per-pixel). Per the CVSS 3.1 vector (AV:L/AC:H/PR:N/UI:R/S:U/C:N/I:H/A:H), exploitation is local-only, requires the victim to actively start the export, requires no privileges (PR:N), and is gated by high attack complexity because the image dimensions must be crafted so the size arithmetic wraps; the assessment rates confidentiality impact as none and integrity/availability impact as high. No public exploit code and no confirmed active exploitation (no CISA KEV listing) were identified at time of analysis, and the EPSS-style signals provided are consistent with a moderate-priority rather than critical issue, since the plug-in is standard GIMP functionality but the trigger depends on user-driven export of deliberately large images.
Stored and reflected cross-site scripting in Gophish through 0.12.1 allows an attacker who controls or intercepts the SMTP server configured in a sending profile to inject JavaScript that runs in an administrator's browser when campaign results are viewed or a test email is sent, enabling theft of the administrator's API key. Exploitation is unauthenticated from the attacker's perspective (PR:N) but requires a significant precondition: control over, or man-in-the-middle interception of, the sending profile's SMTP endpoint, plus passive interaction by a privileged administrator; the independent assessment rates this as a low-priority issue with a CVSS:3.1 score reflecting high attack complexity (AC:H), required user interaction (UI:R), and scope change (S:C) with only low confidentiality and integrity impact. Publicly available exploit code exists, but there is no indication of confirmed active exploitation (no CISA KEV listing).
Concurrent HTTP POST requests to MISP's email OTP login endpoint can both authenticate using the same one-time password, breaking the single-use guarantee and establishing duplicate sessions. Affected versions before 2.5.48 are vulnerable when email OTP login is enabled; exploitation requires the attacker to already hold a valid, unexpired OTP (via interception or social engineering) and to time two requests within the narrow race window. No public exploit code or active exploitation has been identified, and the independent assessment rates this as a genuine but low-urgency integrity issue (CVSS 3.1 base 3.6, high attack complexity) because success merely duplicates a session the OTP already granted.
Integer overflow in GIMP's GIF export plug-in leads to a heap-based buffer overflow when a user exports an image whose width and height are large enough to wrap the 32-bit pixel-buffer size calculation, resulting in an out-of-bounds write that can crash the application or corrupt memory. Affected software is the GIMP package as distributed in Red Hat Enterprise Linux 6, 7, 8, 9 and 10, and exploitation is local (AV:L) with high attack complexity (AC:H) and mandatory user interaction (UI:R) while requiring no privileges (PR:N) - the attacker must induce a user to export an image with extreme dimensions. No public exploit identified at time of analysis, and the independent assessment rates this a moderate-severity, genuinely low-urgency issue rather than a top priority.
Local denial of service in m17n-lib can crash applications on Red Hat Enterprise Linux 6 through 10 when they process specially crafted or truncated UTF-8 input. The flaw is a NULL pointer dereference reachable with a local attack vector and no authentication or user interaction, and it only affects availability, not confidentiality or integrity. No public exploit or confirmed active exploitation has been identified at time of analysis, and the issue is assessed as a moderate-severity local DoS rather than a high-priority remote threat.
Heap use-after-free in m17n-lib's input-method handling allows a local user who supplies specially crafted text to an application using m17n-based input engines to crash that application, producing a denial of service; arbitrary code execution is described as a potential, unverified outcome rather than a demonstrated one. The flaw is rated CVSS 6.2 (CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H), so exploitation is local with no privileges required (PR:N) and no user interaction, and the only confirmed impact is availability loss with no confidentiality or integrity effect. Red Hat tracks the issue for Red Hat Enterprise Linux 6 through 10, and no public exploit code or confirmed active exploitation was identified at time of analysis.
Sustained CPU consumption in m17n-lib allows an unauthenticated local attacker to hang applications that parse attacker-influenced text, because a crafted invalid UTF-8 byte sequence drives the library's text-parsing routine into a non-terminating loop (CWE-835) and pins a CPU core indefinitely, producing a denial of service (CVSS 6.2, AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H). Only systems where an application actually routes externally supplied text through m17n-lib's parsing path are exposed; hosts that never load the library are unaffected, and there is no code execution, data leakage, or integrity impact. There is no public exploit identified at time of analysis, and no vendor-released patch identified at time of analysis; Red Hat is tracking the issue under CVE-2026-107168 with Bugzilla 2547409.
Cross-project queue access in OpenStack Zaqar's WebSocket transport lets an authenticated user who holds a valid Keystone token for one project substitute a different project's UUID in subsequent requests, then enumerate, inspect, create, or delete queues belonging to that other tenant. Only deployments that deliberately enable the WebSocket transport together with Keystone authentication are exposed; the more widely used REST transport and non-Keystone setups are explicitly out of scope. Publicly available exploit code exists (SSVC exploitation: poc), the vendor has released a fix in 23.0.1, and the assessed impact is low confidentiality but high integrity and availability loss within the targeted project.
Reflected cross-site scripting in the Fanvil x7a IP phone firmware (2.6.0.1182) allows a remote, unauthenticated attacker to inject HTML and JavaScript into the device log component, which then executes in the browser of any user who opens the affected log view - in practice an administrator using the phone's web management interface. The CVSS 3.1 base score is 6.1 (AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N), reflecting that exploitation requires user interaction (the victim must render the log page) while the injected script runs in a scope beyond the vulnerable component, though confidentiality and integrity impact is limited to the viewer's browser session rather than the underlying device OS. No public exploit code has been identified at time of analysis and the issue is not listed on CISA KEV, so risk is currently tied to exposure of the management interface rather than observed exploitation.
Reflected cross-site scripting in the Fanvil X7A IP phone's device log component (firmware 2.6.0.1182) lets an attacker inject HTML into log data that the web interface renders without sanitization, so malicious JavaScript executes in the browser of whoever views that log page. The attack is network-reachable but requires user interaction (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N, 6.1), and in practice an operator likely has to open the log view from the phone's administrative web interface, which narrows real-world exposure. Risk signals are low: EPSS is 0.18% (7th percentile), there is no CISA KEV entry, and no public exploit code has been identified at time of analysis - only a third-party vulnerability disclosure writeup.
Stored cross-site scripting in the Aurora Heatmap WordPress plugin affects all versions up to and including 1.7.2 and lets unauthenticated attackers inject arbitrary JavaScript through the plugin's 'url' parameter, with the payload firing in an administrator's browser whenever that admin clicks the injected heatmap link. The flaw is rated CVSS 3.1 6.1 (AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N), meaning no authentication is needed to plant the payload but a successful attack depends on an administrator later interacting with the malicious link; impact is limited to information disclosure and limited integrity within the admin's browser session, with no availability impact. There is no public exploit identified at time of analysis, and the plugin should be treated as untrusted until a fixed build is confirmed. An upstream Trac changeset (3712276) indicates code changes addressing the issue, but no patched release version has been independently confirmed.
Stored cross-site scripting in the SonicWall SMA1000 Appliance Management Console (AMC) allows an attacker who already holds administrator credentials to persist arbitrary JavaScript that then executes in the browser session of another privileged AMC user under the vendor's stated 'specific conditions.' The flaw affects SMA1000 12.4.3-03526 (platform-hotfix) and 12.5.0-02952 (platform-hotfix) along with all older builds, and is rated moderate severity at CVSS 6.1 with low confidentiality/integrity impact and no availability impact. The only hard gate is authentication: an independent assessment places the requirement at PR:H, meaning the attacker must already be an AMC administrator, and exploitation also requires a victim to view the poisoned page (UI:R). No public exploit code has been identified, CISA's SSVC framework records exploitation as none and automation as not feasible, and the available signals converge on LOW real-world priority rather than an urgent threat.
Authenticated Backup Viewer accounts in Veeam Backup & Replication can modify the Enterprise Manager master key and stored antivirus-update credentials due to missing authorization (CWE-862), affecting deployments with Enterprise Manager and that low-privilege role. The issue is medium severity (CVSS 4.0 score 6.1) and requires valid Backup Viewer credentials plus an additional attack precondition, so it is not anonymous or unauthenticated exploitation; no public exploit identified at time of analysis. Veeam KB4934 is the referenced advisory for affected and fixed builds.
Persistent denial of service in Absolute Secure Access servers prior to 14.60 allows authenticated attackers to exhaust proxy sub-system resources by sending malformed data, requiring a valid session and an additional environmental precondition outside the attacker's direct control. The flaw is a CWE-400 uncontrolled resource consumption issue with impact limited to availability; confidentiality and integrity are unaffected. No public exploit code has been identified at time of analysis, and the vendor has released a patch in version 14.60.
Missing authentication in the smart_proxy_dynflow remote-execution callback handlers lets an unauthenticated network attacker forge progress and completion reports for a running job, causing the job to be recorded as a success or a failure with attacker-supplied output. The flaw (CWE-306) applies only when Red Hat Satellite 6 remote execution is configured in pull or pull-mqtt mode, and the attacker must already know the identifier of a live job and submit the forged report during that job's execution window, which is why the assessed severity is moderate (CVSS 5.9, AC:H, integrity-only). No public exploit code and no confirmed active exploitation were identified at time of analysis.
Cleartext database credential disclosure via man-in-the-middle interception affects pgjdbc (PostgreSQL JDBC Driver) versions 42.7.11 through 42.7.13 when the caller-supplied requireAuth connection property is set to exclude all six authentication methods the driver recognizes (for example requireAuth=!password,!md5,!gss,!sspi,!scram-sha-256,!none) or to a value containing no method at all (requireAuth=, a single comma). In that state the driver fails open and accepts whatever method the server requests, including cleartext password authentication, so an attacker who already holds an active man-in-the-middle position between the application and the PostgreSQL server can request cleartext authentication and capture the database password. Exposure is narrow: requireAuth has no default value, so deployments that never set the property are unaffected, and positive lists (requireAuth=scram-sha-256) or partial exclusions (requireAuth=!password,!md5) are enforced correctly; no public exploit identified at time of analysis and the issue is not in CISA KEV. Scoring is confidentiality-only at CVSS 5.9 with high attack complexity (AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N), reflecting that both a specific client misconfiguration and a pre-existing network foothold are required.
Proxied plaintext WebSocket (ws://) requests issued through AsyncHttpClient before 3.0.12 or 2.16.1 leak the configured proxy's credentials to the origin WebSocket server. When an HTTP proxy requiring authentication is in use, the tunneled ws:// upgrade is built with the proxy's Proxy-Authorization header and an absolute-form request target, so the origin endpoint operator - or any party able to observe origin-bound traffic - can recover Basic credentials directly or capture a Digest challenge-response for offline cracking or replay. Exploitation depends on a non-default configuration (an authenticated HTTP proxy plus an application-initiated ws:// request), yields only proxy credentials rather than user or application data, and is rated CVSS 3.1 base 5.9; no public exploit code and no confirmed active exploitation have been identified at time of analysis.
HashiCorp Vault's PKI secrets engine ACME server, when running under the default directory policy, can issue certificates that contain identity claims the ACME challenges never validated, allowing an unauthenticated ACME client (CVSS vector PR:N/UI:N) to obtain a certificate for an identity it does not control and impersonate that identity toward systems that trust the affected Vault PKI mount. The issue is a configuration-dependent certificate-integrity flaw (CVSS 5.9, AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:H/A:N) rather than an urgent remote-code or data-exposure bug: it requires a PKI mount with the ACME server enabled under the default directory policy and relying systems that actually trust certificates issued by that mount, otherwise the unverified identity has no effect. HashiCorp addressed it in Vault Community Edition 2.1.2 and Vault Enterprise 2.1.2, 1.21.12, 1.20.17, and 1.19.23; no public exploit code and no confirmed active exploitation were identified at time of analysis, and no EPSS score was supplied in the available data.
Unauthenticated remote attackers can temporarily disrupt Cisco NX-OS devices by flooding a data-plane interface with a high rate of UDP or TCP connections, exhausting resources and causing packet loss that destabilizes routing and control-plane protocols. The denial-of-service condition is availability-only and self-clears once the traffic stops, but the CVSS 3.1 score of 5.8 reflects a scope change and low availability impact. No public exploit code or active exploitation has been identified at time of analysis.
EPG contract bypass in Cisco Nexus 9000 Series Fabric Switches running ACI Mode allows an unauthenticated, remote attacker to evade configured endpoint-group segmentation by sending IPv4 or IPv6 UDP packets with DHCP source and destination ports (67/68) through the device. The flaw is an improper access-control implementation (CWE-284) rated CVSS 5.8 with scope change and low integrity impact; no public exploit identified at time of analysis. Exploitation requires ACI mode and EPG contracts in use, and the attacker must have a network path through the switch; it does not directly disclose data or cause denial of service.
Path traversal in Quasar App Vite 3.1.0 through 3.2.0 allows a local attacker with low privileges (PR:L per the assessed CVSS vector) who can influence SSG page definitions to make `quasar build -m ssg` create new files or directories outside `build.distDir`; the assessed vector is CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:N. Default deployments are not normally exploitable because Quasar configuration and SSG page definitions are trusted input, and the SSG renderer's no-overwrite behavior limits impact to new files rather than modification of existing files. No KEV status or public exploit was identified at time of analysis, and this is not remotely reachable against a running service; the vendor patch fixes the issue in @quasar/app-vite 3.3.0.
Memory exhaustion in the SSSD Kerberos Credential Manager (sssd-kcm) responder lets a local user or process that can reach the KCM UNIX domain socket deny service to credential caching. An attacker sends a request with an oversized length header and then stalls the connection, causing the responder to preallocate large buffers for each such request until its memory is depleted. Impact is limited to availability (CVSS 5.5, AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H) on hosts where the kcm responder is enabled and its socket is reachable; EPSS is very low (0.18%, 7th percentile) and there is no public exploit identified at time of analysis.
Integer overflow in ImageMagick's pixel-cache allocation path (GetVirtualPixels / SetPixelCacheNexusPixels) lets a locally supplied crafted call with attacker-controlled width, height, and channel/metacontent values overflow MagickSizeType arithmetic and write past a heap buffer, crashing the affected process. All ImageMagick releases prior to 6.9.13-56 (6.x line) and 7.1.2-31 (7.x line) are affected; the assessed CVSS is 5.1 (AV:L/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H), meaning impact is confined to a denial of service with no confidentiality or integrity loss and no demonstrated code execution. Exploitation is local and high-complexity because a precise overflow condition must be induced; there is no CISA KEV listing and no public exploit code identified at time of analysis, and EPSS data was not provided in the available intelligence. Vendor-released patches exist in 6.9.13-56 and 7.1.2-31.
Out-of-bounds heap reads in GIMP's XMC (X cursor) thumbnail loader can crash the application or corrupt process memory when a victim opens a crafted XMC file, or when thumbnail generation is triggered for one by a file manager. The root cause is an integer overflow: the loader sizes a pixel buffer using width * height computed in 32-bit signed arithmetic, so an oversized product wraps to a small allocation that a later GEGL buffer read then indexes using the unwrapped dimensions. Exploitation is strictly local and user-interaction dependent (CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H, 5.5), with no network attack surface and no confidentiality or integrity impact; no public exploit code and no confirmed active exploitation were identified at time of analysis, and Red Hat tracks the issue across RHEL 6 through 10.
Path traversal in VisiData's EmailSheet extract_parts functionality lets a specially crafted .eml file write files to arbitrary locations on the filesystem using attacker-controlled filenames, producing a high integrity impact (C:N/I:H/A:N) on data the user account can write. Only the development snapshot identified by commit 38b21f78 is confirmed affected, so exposure is limited to users running that specific dev build rather than a numbered stable release. The CVSS 3.1 vector (5.5, AV:L/AC:L/PR:N/UI:R) describes a local attack with no privileges required but a mandatory user-interaction step: the victim must open the malicious message and invoke extract_parts, and accordingly there is no remote or automated trigger. At time of analysis there is no evidence of confirmed active exploitation (not listed in CISA KEV) and no public exploit code identified, making this a moderate, low-priority issue.
A use-after-free in the XInput2 passive grab handling of X.Org X server builds prior to xorg-server-21.1.25 and Xwayland prior to xwayland-24.1.14 allows an authenticated client that can add and remove XI2 master devices to crash the server or potentially achieve arbitrary code execution. The attacker creates a passive grab whose modifierDevice points at a master device, then calls XIRemoveMaster to free that device; the grab retains a dangling DeviceIntPtr that later input processing dereferences, corrupting the heap. Reaching the vulnerable code requires a valid client connection to the X server or Xwayland (typically local, since modern X servers do not listen on TCP by default) and the XI2 extension, which is enabled by default; no public exploit code and no CISA KEV entry have been identified at time of analysis.
Information disclosure in iTerm2 for macOS before 3.6.12 allows a local, unprivileged attacker to man-in-the-middle the terminal's mono-server session-restoration handshake and obtain sensitive session data (CVSS 3.1 base 5.5, CWE-200). The flaw lives in iTermFileDescriptorClientConnect(), which connects to a predictably named rendezvous socket (iTerm2.socket.<pid>) inside the world-writable, sticky shared temp directory (/var/tmp); an attacker who pre-creates that socket and wins a timing race can pass an attacker-controlled pty master back to the client via SCM_RIGHTS and read/interpose the restored session. Exploitation requires a local account on the same host, session restoration to be triggered, pid prediction and a race to be won, which is why this is a low-to-moderate priority rather than an urgent one; no public exploit has been identified at time of analysis and CISA SSVC records Exploitation: none, Automatable: no, Technical Impact: partial (the flaw is not in CISA KEV). The vendor fix is included in iTerm2 3.6.12.
Cross-namespace privilege escalation in HashiCorp Vault Enterprise allows an authenticated token holding a specially crafted policy name containing path-traversal constructs to have its policy resolved against a different namespace - including the root namespace - thereby gaining capabilities granted by policies it should never reach. Only Vault Enterprise deployments with namespaces actually configured are exposed; Vault Community Edition does not implement namespaces and is explicitly not affected. Exploitation requires a valid token (CVSS PR:L, authenticated) and the resulting privilege gain is bounded by whatever policies exist in the traversed namespace, making this a moderate-priority privilege-escalation issue rather than a mass-exploitation threat - no public exploit code has been identified at time of analysis and no EPSS or KEV data was supplied.
Stored cross-site scripting (CWE-79) in the Apache Impala Web UI query-plan view allows an authenticated SQL user holding only SELECT permission to embed JavaScript in a table alias, which then executes in the browser of any other user who opens that query plan. All Impala releases from 2.7.0 through 4.5.2 are affected; the vendor-released fix is version 4.5.3. Exploitation requires a valid Impala account plus a second user opening the crafted plan (CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N), and no public exploit identified at time of analysis. Risk is concentrated in multi-tenant clusters where low-trust analysts share an environment with administrators, since the injected script runs in the victim's browser origin rather than against the Impala backend.
Authenticated attackers holding only Subscriber-level access can tamper with the personal records of arbitrary LatePoint customers on WordPress sites running the Appointment Booking & Reservation plugin through version 5.6.9. The process_step_customer() handler merges attacker-supplied POST data into an existing customer record while checking only is_user_logged_in(), so any logged-in account can rewrite another customer's first name, last name, email, phone, and notes; where the non-default contact_merge setting is 'phone', this escalates to overwriting the victim's email and enabling account takeover via password reset. Exploitation requires a valid low-privilege session (PR:L), which is trivial on sites with open registration, and there is no public exploit identified at time of analysis, though the EPSS-style risk is moderate rather than critical.
wger before 2.6 contains a CSRF bypass (CWE-352) that lets an attacker-controlled web page silently rebind an authenticated trainer's session to an arbitrary user account. Because the trainer_login view (wger/core/views/user.py, ~lines 161-210) accepts GET and calls django_login() directly, it falls outside Django's CsrfViewMiddleware enforcement, which only checks unsafe methods; a single <img> tag pointing at /en/user/<user_pk>/trainer-login forces the victim's browser to issue the request with the session cookie attached. Exploitation is authenticated (CVSS PR:L) and requires user interaction (UI:R) - the trainer must hold trainer privileges and be lured onto the malicious page - and, per the assessed vector, it depends on a non-trivial SameSite=Lax bypass via the view's ?next= 302 redirect chain, which raises attack complexity to AC:H and makes this a targeted rather than opportunistic or wormable issue. No CISA KEV listing or independent public exploit code was identified at time of analysis, though the vendor advisory documents a Playwright-verified reproduction in Chromium 147.
Uncontrolled recursion in ImageMagick's CALS decoder lets a crafted CALS image exhaust the call stack and terminate the decoding process, giving remote unauthenticated attackers a low-impact denial of service (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L). All ImageMagick releases prior to 7.1.2-30 (7.x branch) and prior to 6.9.13-55 (6.x branch) are affected; the vulnerable code path is coders/cals.c (ReadCALSImage). No public exploit code or confirmed active exploitation has been identified at time of analysis, and the practical reach is limited because the target must actually accept and decode the niche CALS Type 1 raster format rather than common JPEG/PNG inputs.
Denial of service in ImageMagick's Adobe Swatch Exchange (ASE) decoder allows a crafted ASE file to crash the process or trigger a long-running, resource-draining operation in all builds prior to 7.1.2-32. The flaw is reachable by unauthenticated remote attackers (CVSS:3.1/AV:N/AC:L/PR:N/UI:N) wherever an application passes untrusted image data to ImageMagick with the ASE coder enabled, but impact is limited to availability (C:N/I:N/A:L) with no confidentiality or integrity effect. There is no public exploit identified at time of analysis and the issue is not in CISA KEV; a vendor patch exists in 7.1.2-32, so the practical risk is moderate and primarily relevant to image-processing services, upload pipelines, and thumbnailers that accept arbitrary files.
Denial of service in ImageMagick caused by an infinite loop when parsing XMP metadata: a crafted XMP profile embedded in an image drives the internal XML entity parser into a non-terminating loop, so the image processing operation never completes and the worker thread or process burns CPU indefinitely. All builds prior to 7.1.2-32 (7.x line) and 6.9.13-57 (6.x line) are affected, including the many applications and language bindings that embed ImageMagick for thumbnail generation, upload validation, or media pipelines. The flaw is remotely triggerable and unauthenticated (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L, score 5.3) wherever an attacker-supplied image reaches an automated parser; impact is limited to availability (hang/resource exhaustion) with no memory corruption, data disclosure, or code execution, and no public exploit code was identified at time of analysis.
Repeated processing of crafted Magick scripts in ImageMagick versions prior to 7.1.2-31 leaks a file descriptor on each invocation, allowing unauthenticated remote attackers to exhaust a process's open-file-descriptor table and deny service to applications that exercise the MagickWand script-token (MSL/conjure) path. The issue is rated CVSS 5.3 with availability-only impact (C:N/I:N/A:L) and requires sustained or repeated request volume to be effective, since a single invocation is harmless and the leak is cleared by a process restart. No public exploit identified at time of analysis, and the vendor-released patch is version 7.1.2-31.
Denial of service in ImageMagick versions prior to 7.1.2-32 allows remote, unauthenticated attackers to crash the process by supplying a crafted EXR image that triggers a heap buffer overflow in the EXR decoder. The vulnerability requires an ImageMagick build with EXR/OpenEXR delegate support enabled and an application that passes attacker-controlled EXR files to it; the impact is limited to a single-process crash (low availability). No public exploit code has been identified at time of analysis, and the vendor has released a patch in version 7.1.2-32.
An authenticated remote attacker with a valid SSH session can trigger a one-byte out-of-bounds stack write in wolfSSH v1.4.11 through v1.5.0 on non-Windows platforms by sending a crafted SFTP path, which can corrupt an adjacent stack value and crash the server process. The flaw is an unsigned integer underflow (CWE-191) in which the pointer/length math in wolfSSH_RealPath() and wstrncat() wraps once the accumulated path reaches roughly half the output buffer, causing a terminating null byte to be written one byte past the end of the stack buffer. The impact is limited to a single-byte corruption causing denial of service rather than arbitrary write or code execution, and no public exploit has been identified at time of analysis; a separate, more severe unbounded-copy condition exists only for third-party applications that call the public wolfSSH_RealPath() API with an output buffer smaller than the input path.
Resource exhaustion in ImageMagick's MAT image decoder (coders/mat.c) lets a crafted, highly compressed MAT file inflate far beyond the configured security-policy resource limit, writing an oversized temporary file to disk and consuming excessive storage until the host runs out of temp space. All ImageMagick releases before 7.1.2-30 (7.x line) and before 6.9.13-55 (6.x line) are affected, and the flaw is unauthenticated and remotely reachable per the assessed vector (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L, base 5.3) wherever untrusted image input is accepted, such as upload, conversion, or thumbnailing services; deployments that only process trusted local images face a much narrower exposure. Impact is confined to availability - there is no code execution, data disclosure, or integrity loss - and the vendor has shipped fixes in 7.1.2-30 and 6.9.13-55; no public exploit has been identified at time of analysis.
Denial-of-service in ImageMagick versions before 7.1.2-30 and 6.9.13-55 allows remote unauthenticated attackers to terminate an image-processing process by supplying a crafted image containing a deeply recursive XMP profile. The flaw only affects availability of the single ImageMagick task or worker-CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L-because the recursion limit is treated as a fatal error rather than a recoverable exception; there is no disclosure, corruption, or code execution. At time of analysis, no public exploit code or active exploitation was identified.
Formula evaluation in the Excelize Go library (versions 2.10.1 through 2.11.0) can be crashed by a spreadsheet whose cells contain a RIGHT() formula applied to supplementary-plane (non-BMP) text, because the function validates the requested length in UTF-16 code units but then slices a rune array using code-point counts, producing a negative slice index and a Go panic. This is a denial-of-service (availability-only) issue: per the assessed vector CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L, there is no confidentiality or integrity impact, and remote unauthenticated attackers can trigger it only if the host application calls Excelize's CalcCellValue on attacker-controlled workbook content. No public exploit code was identified at time of analysis, and although an upstream fix commit and pull request exist, no released patched version has been confirmed.
Splunk MCP Server versions below 1.2.1 transmit the Splunk platform authentication token of the user who executes a custom API tool to whatever URL that tool is configured to call, so an attacker who can author a custom tool pointing at a server they control can capture the victim's token and then act as that user. Exploitation is authenticated and requires user interaction: the attacker needs privileges to create or configure a custom API tool, and a separate victim holding a role with the mcp_tool_execute capability must actively run that tool (CVSS:3.1/AV:N/AC:H/PR:H/UI:R/S:U/C:H/I:L/A:L). Real-world risk is moderate - it does not affect default deployments that make no use of custom API tools, and no public exploit has been identified at time of analysis. A vendor patch is available in Splunk MCP Server 1.2.1.
Improper access control in Splunk Enterprise (versions prior to 10.4.3, 10.2.7, 10.0.10, and 9.4.15) lets unauthenticated, network-positioned attackers reach functionality or data that should be restricted, exposing limited sensitive information with no integrity or availability impact (CVSS 3.1 base 5.3, AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N). The flaw is tracked as CWE-284 and was fixed by Splunk as part of a set of internally identified issues grouped by weakness class, each receiving its own CVE; no public exploit code has been identified at time of analysis, EPSS is a low 0.15% (3rd percentile), and CISA's SSVC assessment rates exploitation as none with partial technical impact. Because exploitation requires only network reachability of the affected interface and no credentials or user interaction, the practical exposure is driven mainly by whether a Splunk web/management interface is exposed beyond a trusted network segment.
Residual-data disclosure in the PostgreSQL JDBC Driver (pgjdbc) 42.7.4 through 42.7.13 allows an authenticated attacker who can influence how an application writes a value to recover SQL text and parameter values belonging to earlier statements on the same connection. When an application declares a value length larger than the bytes it actually supplies through PreparedStatement.setObject with a ByteStreamWriter, CopyIn.writeToCopy, PGCopyOutputStream.write, LargeObject.write, or Blob.setBytes, the driver pads the shortfall with whatever bytes remain in its send buffer instead of zeros, exposing up to 8192 bytes of recent traffic per value (16320 bytes when GSS encryption is in use); on a pooled connection those bytes may belong to other tenants or requests. Exploitation is conditional rather than trivial: the affected application must both store and read back such a padded value, its declared lengths must not always match its data, and any cross-request leakage depends on connection pooling, which is why the vendor rates this 5.3 (CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:N/A:N). No public exploit code was identified at time of analysis, and there is no confirmed active exploitation (not in CISA KEV).
Silent threat-intelligence data loss occurs in MISP deployments that federate objects: when a user with object-creation privileges saves a MISP Object with an empty description, the originating instance accepts it, but the receiving instance during synchronization rejects it and silently drops the object together with all of its associated attributes. The flaw affects versions below 2.5.48 and requires an authenticated object creator (CVSS 4.0 PR:L) in a multi-instance sync topology, so it is a low-urgency data-integrity defect rather than an externally exploitable breach; no public exploit identified at time of analysis, and the root cause is fixed upstream by the validation change in MISP commit 6b1776f07.
MISP installations prior to 2.5.48 fail to refresh event correlations after an authenticated user with write access changes an event's distribution level or sharing group, because the before-save hook recorded the incoming (new) values instead of the already-persisted ones and the subsequent after-save comparison therefore found no change to act on. The practical effect is unintended information disclosure: correlations computed under a broader sharing scope can remain visible to members of the previously wider group even after the event has been moved into a more restrictive one, while distribution widening can conversely leave newly eligible correlations missing. Exploitation requires an authenticated user (CVSS PR:L) who can edit at least one existing event and who touches one of those two fields; no public exploit code has been identified at time of analysis and no CISA KEV entry exists, so this is a genuine-but-low-priority access control defect rather than an actively exploited flaw.
Authorization bypass in Obot 0.26.0 through 0.26.1 lets an authenticated user who matches any configured vMCP (virtual MCP) profile selector read prompts, resources, and resource templates belonging to components that profile never granted them, because profile enforcement in those releases was applied only to tools. Attackers abuse the vMCP owner's shared component connection to reach the ungranted prompt/resource endpoints over the network, yielding limited confidentiality impact with no integrity or availability effect (CVSS 5.3, CWE-863, PR:L, so valid authentication and a profile-segmented vMCP deployment are prerequisites). There is no public exploit identified at time of analysis and no confirmed active exploitation (CISA KEV); vendor-released patch v0.26.2 closes the gap.