Suse
Monthly
Cross-origin data leakage via Chrome's Autofill implementation on Android allows a remote attacker with prior renderer process compromise to extract sensitive data across origin boundaries using a crafted HTML page. Affected versions are all Chrome for Android releases prior to 150.0.7871.47. No public exploit code has been identified and the EPSS score of 0.21% (11th percentile) reflects low observed exploitation activity; however, the cross-origin data exposure (CWE-346) combined with Autofill access creates meaningful privacy risk for users on vulnerable Android builds. Google has released a patch in the stable channel update.
Navigation restriction bypass in Google Chrome's Passwords subsystem allows a remote attacker to circumvent intended access controls by delivering a crafted HTML page to a user. All Chrome desktop versions prior to 150.0.7871.47 are affected, with integrity rated High by CVSS due to the ability to manipulate navigation behavior. No public exploit exists at time of analysis and EPSS sits at 0.22% (13th percentile), indicating low near-term exploitation probability despite Chrome's enormous global footprint and Chromium's internal 'High' severity classification.
Cross-origin data leakage in Google Chrome for Android prior to 150.0.7871.47 enables remote attackers to read data from other origins via a crafted HTML page exploiting insufficient validation in the File Input component. The flaw is Android-platform-specific, requiring only that a victim user navigate to an attacker-controlled page. No active exploitation has been confirmed (CISA KEV negative, SSVC exploitation: none), and EPSS at 0.21% (11th percentile) reflects low current exploitation interest, though the unauthenticated, low-complexity attack vector keeps it a relevant patching priority for Android Chrome deployments.
Universal Cross-Site Scripting (UXSS) in Google Chrome for iOS prior to 150.0.7871.47 enables a remote attacker to inject arbitrary scripts or HTML across origins by exploiting insufficient input validation triggered through specific user UI gestures on a crafted page. The scope change (S:C in CVSS) is the critical dimension here - successful exploitation bypasses the same-origin policy, potentially granting the attacker script execution in the context of arbitrary origins within the browser session. No public exploit code has been identified and EPSS sits at just 0.20% (11th percentile), indicating low observed exploitation activity despite the high-impact class of vulnerability.
Memory disclosure in Google Chrome for Linux (prior to 150.0.7871.47) allows a remote attacker to read sensitive contents from the browser's process memory by directing a user to a crafted HTML page. The flaw stems from an inappropriate implementation in Chrome's input handling subsystem, and is platform-specific - only Linux builds are affected. No public exploit code or CISA KEV listing has been identified at the time of analysis, but the network-reachable, zero-authentication attack path and high confidentiality impact make patching a near-term priority for Linux desktop environments.
Sensitive process memory exposure in Google Chrome for iOS prior to 150.0.7871.47 enables a physically present attacker to read potentially sensitive data from the browser's memory without authentication. The root cause is insufficient data validation (CWE-20) within the iOS-specific Chrome implementation. No public exploit code or CISA KEV listing has been identified at time of analysis, and the physical access requirement (CVSS AV:P) significantly constrains real-world attacker opportunity.
Navigation restriction bypass in Google Chrome for iOS (prior to 150.0.7871.47) permits a remote, unauthenticated attacker to circumvent client-side policy controls by delivering a crafted HTML page to a victim user. The flaw originates from insufficient policy enforcement (CWE-602), where navigation security logic enforced within the browser client can be subverted through malicious web content. No active exploitation has been identified - CISA KEV is not listed, SSVC confirms Exploitation: none, and the EPSS score of 0.22% (13th percentile) reflects minimal real-world exploitation activity at time of analysis.
Cross-origin data leakage in Google Chrome's SVG rendering subsystem (versions prior to 150.0.7871.47) allows remote attackers to exfiltrate sensitive information from other origins by directing victims to a crafted HTML page. The root cause is insufficient policy enforcement in SVG handling, classified as CWE-346 (Origin Validation Error), effectively bypassing Same-Origin Policy protections. No public exploit has been identified at time of analysis, and EPSS places exploitation probability at 0.21% (11th percentile), indicating low near-term exploitation likelihood despite a CVSS confidentiality impact rated High.
Side-channel information leakage via Chrome's Scroll implementation exposes cross-origin data to remote attackers who can lure a victim to a crafted HTML page. All Chrome versions prior to 150.0.7871.47 are affected; exploitation requires user interaction (visiting the attacker-controlled page) but no authentication. No public exploit identified at time of analysis and EPSS sits at the 11th percentile, though the confidentiality impact is rated High by NVD given the potential to read data from foreign origins.
ImageMagick before 7.1.2-24 fails to correctly enforce path restrictions configured in its security policy (policy.xml), enabling file creation and truncation outside designated filesystem boundaries. In sandboxed image conversion services - a common deployment pattern for web applications accepting user-submitted images - this incorrect policy check can be exploited by attackers who submit crafted images to write arbitrary files accessible to the conversion process. No public exploit has been identified at time of analysis and this is not listed in CISA KEV, but the wide use of ImageMagick in web application stacks makes the affected population substantial.
Nonce reuse in ImageMagick's AES-CTR cipher implementation exposes encrypted image plaintext to recovery attacks. The PasskeyEncipherImage method in ImageMagick before 7.1.2-22 reuses nonces when performing AES in Counter mode, violating the fundamental security requirement that a nonce be used exactly once per key. A network-accessible attacker who can collect multiple ciphertexts produced with the same passkey and nonce can XOR the outputs to cancel the shared keystream and partially or fully recover encrypted image content. No active exploitation has been identified (not in CISA KEV), and no public exploit code is known at time of analysis.
Memory exhaustion via denial of service in ImageMagick before 7.1.2-19 allows remote attackers to trigger a persistent memory leak in the PNG encoder by supplying crafted MNG image input that induces an encoder failure condition. Affected deployments include any service that accepts user-supplied images and processes them through ImageMagick's MNG write path. No public exploit code or active exploitation has been identified at time of analysis, but the network-reachable attack surface in image-processing pipelines makes this a realistic DoS risk against unpatched instances.
Division by zero in ImageMagick's binomial kernel processing path, affecting all versions before 7.1.2-22, allows an attacker to crash the application by supplying a crafted image with an oversized binomial kernel value. The integer overflow corrupts the kernel size calculation, and a subsequent division by the resulting zero value causes an unhandled exception and process termination. No public exploit code has been identified and the vulnerability is not listed in CISA KEV; however, automated image-processing pipelines that accept untrusted uploads are at elevated practical risk despite the local-vector CVSS rating.
Heap buffer out-of-bounds read in ImageMagick before 7.1.2-19 via off-by-one error in morphology validation allows local attackers with user interaction to trigger denial of service. The vulnerability stems from incorrect morphology parameters causing single pixel memory access violations within heap buffers, potentially leading to application crashes or information disclosure through controlled reads of adjacent heap memory.
### Summary Responsibly disclosed by @NSEcho. Rated medium severity (CVSS 4.9), this vulnerability is remotely exploitable, low attack complexity. This Uncontrolled Resource Consumption vulnerability could allow attackers to cause denial of service by exhausting system resources.
### Impact Shovel and Federation plugins perform URI obfuscation in their worker (link) state. Rated medium severity (CVSS 5.5), this vulnerability is low attack complexity.
Path traversal in Rancher Fleet's ImageScan subsystem (CWE-23) allows authenticated remote attackers to escape intended directory boundaries and trigger denial of service across four active release branches (0.12.x through 0.15.x). The flaw appears to have been introduced in the 0.12.0 series and persisted undetected through at least sixteen subsequent patch releases, indicating the ImageScan component lacked adequate path sanitization for an extended period. No public exploit code or active exploitation has been identified; the CVSS 4.0 vector (PR:L) confirms exploitation requires authenticated access, materially limiting opportunistic attack surface.
Privilege persistence in SUSE Rancher allows project users to retain Pod Security Admission (PSA) permissions even after an administrator revokes those permissions from a RoleTemplate, due to a missing cleanup step in the legacy Project Role Template Binding (PRTB) reconciler. Affected versions are Rancher 2.13.0 through 2.13.7 and 2.14.0 through 2.14.3. An attacker with a pre-existing role assignment can continue to bypass PSA policies enforced at the project level, defeating administrative intent and potentially deploying workloads that violate the cluster's security posture. No public exploit has been identified at time of analysis, and the flaw is not listed in CISA KEV.
Denial of service in fzf's --listen mode HTTP server arises from quadratic-time (O(n²)) body processing caused by repeated string concatenation when reading POST request payloads. All fzf releases prior to 0.73.1 are affected when the --listen flag is explicitly activated to expose the embedded HTTP server. An attacker with access to the listening socket can send a single crafted POST request with many small body segments to monopolize the single-threaded server, blocking all other clients indefinitely. No public exploit identified at time of analysis; vendor-released patch is available in version 0.73.1.
Integer overflow in fzf's FuzzyMatchV2 function causes a non-recoverable process crash (denial of service) when input lines reach approximately 2,200,000 bytes and a search pattern of ~999 bytes is active. The Go runtime detects the resulting invalid slice bounds and immediately terminates the process with a panic - there is no recovery path and no possibility of memory corruption or code execution. No public exploit code has been identified and this CVE is not listed in CISA KEV; the CVSS 4.0 score of 5.6 (AT:P, UI:A) reflects that exploitation requires atypical input volumes and active user or script interaction.
Super-linear algorithmic complexity in the decode-uri-component npm package (versions through 0.4.1) allows unauthenticated network attackers to cause severe CPU exhaustion and block the Node.js event loop by submitting crafted strings with large numbers of percent-encoded tokens. The decode() function splits input on '%' and iteratively invokes decodeComponents() in a pattern producing quadratic or worse time growth - 1,400 '%ab' tokens consume approximately 33 seconds of CPU, rendering single-threaded JavaScript runtimes effectively unresponsive. No public exploit is confirmed at time of analysis (CVSS E:U), but the deterministic algorithmic nature makes payload construction trivial for any attacker with a reachable input channel.
Insufficient data exists to characterize CVE-2026-13606. The sole intelligence signal is a vendor report attributed to Ubuntu, indicating this vulnerability affects a package or component within the Ubuntu ecosystem. No description, CVSS score, vector, CWE classification, or additional references are available - a complete technical characterization is not possible from the provided data.
Improper Authorization (CWE-285) in Apache Tomcat's default servlet allows HTTP method-based and method-omission security constraints to be silently bypassed across all major supported Tomcat branches from 7.0.x through 11.0.x. An attacker can perform HTTP operations - such as PUT or DELETE - that the web.xml security constraint configuration was intended to restrict, potentially enabling unauthorized file upload, modification, or deletion on the default servlet's served content. No active exploitation has been confirmed (not in CISA KEV) and no CVSS score has been published at time of analysis, but the broad version range and ubiquitous deployment footprint of Tomcat make prompt patching a priority.
Replay attack vulnerability in Apache Tomcat's cluster EncryptionInterceptor allows a network-adjacent attacker to retransmit previously captured encrypted inter-node cluster messages, causing receiving nodes to accept and process them as legitimate - potentially corrupting distributed session state or triggering unintended cluster actions. All major supported branches are affected: 11.0.0-M1 through 11.0.22, 10.1.0-M1 through 10.1.55, 9.0.13 through 9.0.18, plus end-of-life branches 8.5.38-8.5.100 and 7.0.100-7.0.109. No public exploit or CISA KEV listing has been identified at time of analysis; however, the breadth of affected versions across all active and legacy branches elevates organizational exposure, particularly for environments running EOL 8.5.x or 7.x with no available patch.
Reflected XSS in Apache Tomcat's bundled 'number guess' example application exposes users of that demo page to script injection across all major Tomcat release lines from 7.0 through 11.0. The flaw resides in a sample JSP/servlet, not the core Tomcat runtime, meaning exploitation depends entirely on the example application being deployed and accessible - a configuration that violates standard production hardening guidance. No public exploit code or active exploitation has been identified at time of analysis; no CVSS vector was assigned by the reporter.
Memory corruption (CWE-119) in Apple Safari's web content processing engine causes an unexpected application crash when rendering maliciously crafted web content. Affected across Safari, iOS 26, iPadOS 26, and macOS Tahoe - all versions prior to 26.5.2. An unauthenticated remote attacker can trigger a denial-of-service against any user who visits a hostile page, though no code execution is indicated by current data. No public exploit code and no CISA KEV listing are identified at time of analysis; exploitation probability (EPSS) was not provided in source data.
Server-side request forgery in Pinpoint APM through version 3.1.0 allows authenticated users with low privileges to register internal or cloud-metadata URLs as alarm webhook targets due to missing URL validation on the webhook registration endpoint. By deliberately triggering alarm threshold breaches, an attacker can coerce the Pinpoint server to issue outbound POST requests to internal hosts, cloud metadata services (such as AWS IMDS at 169.254.169.254), or other non-routable internal infrastructure. No public exploit code or CISA KEV listing has been identified at time of analysis, though the CVSS 4.0 subsequent-system confidentiality impact is rated High, reflecting the potential to enumerate internal services or harvest cloud credentials.
Path traversal in spice-vdagent enables a malicious or compromised SPICE host to write arbitrary files to any location on the guest operating system during file transfer operations. Affecting Red Hat Enterprise Linux versions 6 through 10, the flaw stems from unsanitized filenames supplied by the SPICE host before use by the vdagent process, allowing writes at the privilege level of the logged-in guest user. No public exploit is identified at time of analysis, and exploitation requires prior control of the SPICE host - a high-privilege precondition that substantially constrains realistic risk despite the high integrity impact.
Integer overflow in spice-vdagent's message parsing allows a malicious or compromised SPICE host to crash the guest VM's vdagent daemon by delivering a specially crafted message that triggers a downstream heap buffer overflow. Affected deployments span spice-vdagent as packaged in Red Hat Enterprise Linux 6 through 10, where the daemon runs inside guests handling host-to-guest SPICE protocol communications. No public exploit code or CISA KEV listing exists at time of analysis; exploitation is structurally constrained by the requirement that the attacking SPICE host itself be untrusted or already compromised prior to the attack.
Path traversal in ANTLR4 up to 4.13.2 exposes arbitrary file read via the tokenVocab grammar option handler. The vulnerable function getImportedVocabFile in TokenVocabParser.java fails to sanitize user-controlled grammar option values, allowing an attacker to supply a crafted .g4 grammar file that causes the tool to read files outside the intended working directory. A publicly available proof-of-concept exploit exists (GitHub issue reference), and no vendor patch has been released - the vendor did not respond to disclosure. No active exploitation is confirmed in CISA KEV, but the CVSS 4.0 E:P flag confirms exploit code is public.
Code injection in ANTLR4's Grammar Action Block Handler allows remote attackers to execute arbitrary code by supplying a crafted grammar file processed by the tool. All versions up to and including 4.13.2 are affected via the OutputFile.java code generation pathway. A public proof-of-concept exploit exists on GitHub (no KEV listing), though the vendor has not acknowledged or patched the issue, leaving users without an official fix.
Out-of-bounds reads and a process crash in Nmap through 7.99 can be triggered remotely by any host that is being scanned, or by an on-path attacker, via a crafted IPv6 packet with a truncated extension header. The root cause is an integer underflow in the `ipv6_get_data_primitive` function (libnetutil/netutil.cc) where an insufficiently strict bounds check allows the extension-header walk pointer to advance past the captured packet buffer, causing the remaining-length subtraction to wraparound to a very large value. A publicly available exploit proof-of-concept exists; this vulnerability is not currently listed in the CISA KEV catalog.
HTTP request/response smuggling in nghttpx (the reverse proxy component of nghttp2 through 1.69.0) allows unauthenticated remote attackers to poison shared backend keep-alive connections by crafting an HTTP/1.1 Upgrade request that simultaneously carries a Content-Length header and body. When nghttpx forwards this ambiguous message to a backend and re-adds Connection and Upgrade headers while passing Content-Length verbatim, a backend that resolves the parsing ambiguity in the attacker's favor treats the body as a separate, attacker-controlled HTTP request - enabling cross-client response-queue poisoning. A publicly available proof-of-concept exploit exists; no CISA KEV listing at time of analysis.
NULL pointer dereference in the Linux kernel MANA (Microsoft Azure Network Adapter) driver crashes the host kernel when initializing Virtual Function (VF) devices in VFIO passthrough or nested KVM environments where `pdev->slot` is unpopulated. Systems running affected kernel versions in Azure-style or nested virtualization deployments face deterministic kernel panics and total availability loss on the host. No public exploit code has been identified, and EPSS probability is 0.16% (5th percentile), indicating low opportunistic exploitation risk - though in targeted Azure or nested KVM environments, the crash is straightforwardly reproducible.
Deadlock in the Linux kernel's DSA (Distributed Switch Architecture) subsystem allows a local low-privileged user to hang the kernel network stack by running ethtool against a DSA-configured conduit interface. The DSA subsystem replaces the conduit device's ethtool_ops with aggregating wrappers, but these wrappers redundantly invoke netdev_lock_ops() on a lock already held by the kernel ethtool infrastructure, creating a classic double-lock deadlock. No public exploit exists and EPSS sits at 0.15% (5th percentile), though the issue is trivially reproducible with standard system tooling on affected configurations.
Unbounded NAPI busy-poll loop in the Linux kernel's io_uring subsystem allows a low-privileged local process to spin the kernel indefinitely when no I/O events arrive, triggering soft lockup watchdog complaints and degrading system availability. Affected kernels are those at or after commit 8d0c12a80cdeb80d5e0510e96d38fe551ed8e9b5 (introducing NAPI support in io_uring, Linux 6.9) through the patched stable releases 6.18.33, 7.0.10, and 7.1. No public exploit exists and this is not listed in CISA KEV; EPSS sits at 0.15% (5th percentile), reflecting very low exploitation interest.
Improper input validation in the Linux kernel's nilfs2 filesystem driver allows a local low-privileged user to trigger a kernel WARN_ON and cause a denial of service by submitting a crafted ioctl request with bd_oblocknr set to zero. The zero value collides with the -ENOENT sentinel path in nilfs_ioctl_mark_blocks_dirty(), bypassing the dead block detection logic and passing an invalid block reference into nilfs_bmap_mark(). No active exploitation has been identified (EPSS 0.17%, not in CISA KEV), and patches are available across all actively maintained Linux stable branches.
Spurious WARN_ON_ONCE calls in the Linux kernel's blk-wbt subsystem trigger unnecessary kernel warnings and stack traces under expected failure conditions - memory pressure or a pre-registered writeback throttle. Local low-privileged users can provoke this via ioctl(BLKPG) during MTD partition creation under memory-constrained conditions, generating noise in kernel logs and, on panic_on_warn-enabled kernels, potentially causing a system crash. No public exploit exists and the EPSS score of 0.14% (4th percentile) places this firmly in the low-priority tier; it has not been added to CISA KEV.
NULL pointer dereference in the Linux kernel's mt76/mt7925 WiFi driver crashes the kernel when the 'sta' pointer is dereferenced before a NULL check in mt7925_tx_check_aggr(). Systems with MediaTek mt7925 WiFi hardware running affected kernel versions are exposed to a local denial-of-service via kernel panic. No public exploit exists and EPSS is at the 5th percentile (0.16%), indicating no observed widespread exploitation; patches have been applied across multiple stable kernel branches.
Firmware crash in the Linux kernel's mt76/mt7921 WiFi driver allows a local low-privileged user to cause denial of service by triggering station Association ID (AID) values above the firmware's undocumented limit of 20. Affected hardware includes Mediatek mt7922 chipsets operating in AP mode (IFTYPE_AP). Exploitation requires running a non-standard hostapd configuration that allocates AIDs starting at 65 or higher - stock hostapd is not affected. No public exploit has been identified and EPSS is 0.16% (5th percentile), indicating very low exploitation likelihood.
NULL pointer dereference in the Linux kernel's AMD RAS subsystem crashes systems with AMD GPU hardware. The function `ras_core_ras_interrupt_detected()` in `drm/amd/ras` fails to validate the `ras_core` pointer before dereferencing `ras_core->dev` in an error path, enabling a local low-privilege user to trigger a kernel panic and denial of service. No public exploit code exists and the vulnerability is not listed in CISA KEV; EPSS at 0.14% (4th percentile) confirms negligible current exploitation probability. Upstream fix commits are available in stable kernel trees.
NULL pointer dereference in the Linux kernel's drm/amd/ras subsystem allows a local low-privileged user to crash the kernel on systems with AMD GPUs. The flaw in ras_core_get_utc_second_timestamp() occurs when a NULL ras_core pointer passes a conditional check but is then dereferenced in the subsequent error-printing path, producing a kernel panic. No public exploit exists and EPSS sits at 0.15% (4th percentile), placing this firmly as a low-urgency local denial-of-service hardening issue despite the Availability:High CVSS rating.
Denial of service in the Linux kernel's padata parallel data processing subsystem results from a misplaced CPU hotplug callback that returns an error in a section where error returns are not permitted. Specifically, padata_cpu_dead() is registered below CPUHP_TEARDOWN_CPU, a hotplug state phase where callbacks are not allowed to fail; when the callback returns non-zero, the kernel emits a 'DEAD callback error' warning that can destabilize the system. A local attacker with low-privilege access on a multi-CPU system using padata (e.g., for parallel IPsec or crypto processing) can trigger this condition during CPU offline transitions, resulting in high availability impact. No public exploit is identified and EPSS is 0.16%, indicating low exploitation probability; patches are available across multiple stable kernel branches.
Integer overflow on the RISC-V IOMMU invalidation path in the Linux kernel triggers an infinite loop, causing a local denial of service. A low-privileged local user on an affected RISC-V system with IOMMU hardware can hang the kernel by inducing an invalidation where gather->end equals ULONG_MAX, exploiting the sign-extended page table range supported by RISC-V. EPSS is 0.15% (5th percentile), no active exploitation has been identified, and no public exploit code is known.
Uninitialized memory read in the Linux kernel FUSE filesystem driver causes local denial-of-service conditions. Specifically, `fuse_dentry_revalidate()` in `fs/fuse/dir.c` reads the `->d_time` field of a dentry that was allocated by `__d_alloc()` without initialization, detectable by KMSAN as a use-before-initialize defect. A local authenticated user with access to a FUSE-mounted filesystem can trigger a kernel crash by opening a file through the affected `lookup_open` → `d_revalidate` code path. No public exploit exists and EPSS is 0.15%, indicating minimal real-world exploitation pressure at this time.
Kernel page fault in the Linux kernel's Tegra234 Control Backbone (CBB) driver crashes affected NVIDIA Tegra234-based systems when a cross-fabric error interrupt triggers incorrect register address resolution. The `tegra234_cbb_get_tmo_slv()` function uses the receiving fabric's base address (`cbb->regs`) with register offsets derived from a different fabric's target map, producing an invalid virtual address dereference (`ffff80000954cc00`) and a kernel panic. Exploitation requires local low-privilege access to Tegra234 hardware; no public exploit or POC exists, EPSS sits at the 5th percentile (0.15%), and the vulnerability is not in CISA KEV.
Null pointer dereference in the Linux kernel's pinctrl subsystem allows a local low-privileged attacker to crash the kernel, causing a denial of service. The flaw exists in pinconf_generic_parse_dt_pinmux(), which fails to validate that a 'pinmux' Device Tree property is non-empty before use - an empty property causes the allocator to return an invalid non-NULL pointer, and subsequent memory access triggers a kernel panic. No public exploit code exists and EPSS is 0.15% (5th percentile), reflecting low real-world exploitation probability; exploitation is further constrained to systems with specific pinctrl hardware and Device Tree configurations.
Off-by-one array overflow in the Linux kernel's hvc_iucv driver (IBM z/VM IUCV HVC terminal subsystem) can trigger an out-of-bounds write to hvc_iucv_table[8] when hvc_iucv_devices is set to the maximum value of 8, crashing the kernel. Exploitation requires local access with low privileges on IBM System z systems running z/VM with hvc_iucv configured. No public exploit identified at time of analysis; EPSS is 0.18% (7th percentile), consistent with a niche mainframe-only driver affecting a very limited deployment population.
NULL pointer dereference in the Linux kernel's ps883x USB Type-C retimer driver causes a kernel Oops during device unbind, resulting in a denial of service. The ps883x_retimer_remove() function calls i2c_get_clientdata() to retrieve driver-private state, but i2c_set_clientdata() was never called during probe, leaving the pointer NULL; dereference at offset 0x20 triggers a fatal translation fault. Systems with ps883x retimer hardware running affected kernel versions are vulnerable when a privileged local user unbinds the driver via sysfs. No active exploitation is confirmed (no CISA KEV listing), and the EPSS score of 0.17% (6th percentile) reflects negligible real-world exploitation probability.
Soft CPU lockup in the Linux kernel SCSI generic (sg) driver allows a local user to trigger a denial of service by writing an out-of-range value directly to the def_reserved_size sysfs module parameter, bypassing the validation enforced by sg_proc_write_dressz. Setting def_reserved_size to -1 via /sys/module/sg/parameters/def_reserved_size causes sg_build_reserve to enter a non-terminating allocation loop on the next open() of any /dev/sgX device, hanging the affected CPU core for 26+ seconds until the kernel watchdog fires. No public exploit code has been identified and EPSS probability is very low at 0.18%; patches are confirmed available across all active stable kernel branches.
NULL pointer dereference in the Linux kernel's EIP93 hardware crypto driver triggers a kernel panic during AEAD crypto operations on affected embedded systems. The root cause is a self-contradictory algorithm lookup mask in eip93_hmac_setkey() - CRYPTO_ALG_ASYNC is passed as an exclusion mask, yet EIP93 algorithms are inherently async, so the lookup always returns -ENOENT. This leaves the AEAD security association record with zeroed digest fields; any subsequent crypto operation then dereferences a NULL context pointer, crashing the kernel. No public exploit is identified at time of analysis, and EPSS is very low (0.17%, 6th percentile), reflecting the narrow hardware-specific attack surface.
Null pointer dereference in the Amlogic T7 SoC reset driver crashes the Linux kernel when uninitialized reset operation callbacks are invoked, causing a full system denial of service. Local low-privilege users on systems physically equipped with an Amlogic T7 SoC are affected across kernel versions from the introduction commit through the patched releases 6.18.33, 7.0.10, and 7.1. No public exploit has been identified and EPSS of 0.17% (6th percentile) reflects negligible real-world exploitation interest; this is a hardware-specific stability fix rather than an actively exploited security threat.
NULL pointer dereference in the Linux kernel's Airoha QDMA network driver causes a kernel panic (denial of service) when queue entry list allocation fails during TX queue initialization. Systems running Airoha-based network hardware on affected kernel versions - including Linux 6.19 - are vulnerable when the driver's cleanup path dereferences an uninitialized descriptor array pointer. No public exploit exists and EPSS sits at 0.17% (6th percentile), indicating minimal real-world exploitation activity; no CISA KEV listing.
NULL pointer dereference in the Linux kernel's airoha QDMA network driver crashes the kernel when RX queue initialization fails mid-way through allocation. The flaw exists in airoha_qdma_init_rx_queue() where early ndesc initialization causes the cleanup routine to call netif_napi_del() on NAPI structures never registered via netif_napi_add(), dereferencing a null pointer and causing a local denial of service. No public exploit identified and EPSS at 0.17% (6th percentile) reflects minimal real-world exploitation probability; impact is constrained to systems running Airoha QDMA-based network hardware.
Kernel panic via NULL pointer dereference in the Linux kernel MANA (Microsoft Azure Network Adapter) driver allows a local low-privileged user to crash the host when a power management resume failure leaves the driver in a torn-down state. The mana_remove() function is invoked twice - first during a failed mana_probe() recovery path that nullifies gdma_context and driver_data, and again when the driver is subsequently unbound - dereferencing a NULL gc pointer. No public exploit exists and EPSS is 0.17% (6th percentile), indicating minimal real-world exploitation activity; however, the impact on Azure Linux VM availability is complete (kernel panic).
Improper locking in the Linux kernel's amdgpu DRM driver allows a local low-privileged user to trigger a kernel deadlock or system crash via the AMDGPU_INFO_READ_MMR_REG ioctl. Three distinct concurrency defects exist: inverted lock ordering between the reset semaphore and mm_lock (permitting copy_to_user() under lock), memory allocation while holding the reset semaphore, and use of down_read_trylock() where a blocking wait is required. No public exploit code has been identified and EPSS is 0.17%, placing this firmly in opportunistic-rather-than-targeted risk territory.
Kernel crash via NULL pointer dereference in the ALSA HDA Conexant audio driver affects Linux kernel stable branches from 5.15 through 7.0, exploitable by a local low-privileged user on systems with Conexant HDA audio hardware. The cx_probe() function ignores the error pointer returned by snd_hda_jack_detect_enable_callback() on memory allocation failure, leaving the jack detection callback unregistered and an internal structure uninitialized; a subsequent audio jack event then triggers a kernel panic. No public exploit exists and no CISA KEV listing is present; EPSS is 0.17% (7th percentile), confirming very low real-world exploitation probability.
NULL pointer dereference in the Linux kernel ice driver's ice_reset_all_vfs() function crashes the host kernel when a Virtual Function VSI rebuild fails during concurrent NVM firmware updates on Intel E800-series NICs. Systems with SR-IOV enabled on Intel E800-series adapters running kernel versions from 5.8 through the patched stable releases are affected. An attacker or privileged local process can trigger a kernel panic by initiating a VF reset while nvmupdate64e has placed Admin Queue commands in a transitional failure state, resulting in complete host denial of service. No public exploit exists and EPSS is negligible at 0.17% (7th percentile); CISA KEV listing is absent.
Memory corruption in the arm64 early kernel mapping on Linux allows a local low-privileged user to cause a kernel panic through an off-by-one page-table allocation error introduced since commit 5973a62efa34. Affected systems running 4K page granule on arm64 hardware may have the [data, end) segment overflow into the guard gap page before early_init_stack, corrupting critical early initialization memory. No public exploit exists and EPSS is 0.17% (6th percentile), reflecting this as primarily a kernel stability and reliability concern for arm64 deployments rather than an actively weaponized attack vector.
Silent audit trail corruption in the Linux kernel's capability auditing subsystem allows a local attacker who manipulates inheritable capabilities as a precursor to privilege escalation to have that manipulation go unrecorded in CAPSET audit logs since 2008. The `__audit_log_capset()` function records `cap_effective` into the `cap_pi` (inheritable) field due to a copy-paste error, meaning compliance and forensic systems receive falsified capability state that masks the inheritable-cap modification step of a privilege escalation chain. No public exploit has been identified at time of analysis, and the EPSS score of 0.18% (7th percentile) reflects that exploitation as a standalone attack is not observed in the wild.
Boot-time kernel crash in Linux 6.16+ on AMD IOMMU-equipped systems causes a General Protection Fault when a PCI device whose Bus:Device.Function address is absent from the ACPI IVRS table is encountered during IOMMU initialization. The root cause is a missing bounds check in __rlookup_amd_iommu() that was latent until commit e874c666b15b changed the rlookup_table allocation from a zeroed page-order block (which returned NULL on overrun) to a tight kvcalloc(), causing adjacent slab contents to be dereferenced as a valid struct amd_iommu pointer. The result is a non-recoverable GPF at boot time, confirmed in production on Google Compute Engine ct6e VMs; no public exploit code and no CISA KEV listing exist at time of analysis.
Stack access fault in the x86/kexec purgatory handoff crashes non-kjump kexec operations on affected Linux kernel versions, resulting in a local denial of service. A prior regression removed a 'gratuitous' stack push in the non-kjump path; however, the purgatory code shipped by kexec-tools still reads above its stack top expecting that return address, triggering a memory fault that aborts the kexec reboot or crashes the system. No public exploit exists and EPSS sits at the 6th percentile, placing this firmly in the reliability-regression category rather than an adversarial attack surface.
NULL pointer dereference in the Linux kernel IOMMU subsystem crashes affected systems during PCI device reset when a default domain allocation has previously failed. Local users with low-privilege access on Linux kernel versions in the 7.0 branch (prior to 7.0.10) and before 7.1 can trigger a kernel panic through the `pci_dev_reset_iommu_done()` function, resulting in a denial of service. No public exploit code exists and no active exploitation has been confirmed; EPSS of 0.15% (5th percentile) reflects the low real-world exploitation likelihood given the edge-case hardware precondition and local-only attack vector.
Indefinite kernel hang in the Linux kernel's drm/gma500 Oaktrail LVDS display initialization code can be triggered locally on systems with Intel GMA 500/Oaktrail graphics hardware, causing a denial of service requiring a hard reboot. The defect lies in error handling that incorrectly attempts to deregister an I2C adapter obtained via i2c_get_adapter() - which only increments a reference count - rather than restricting cleanup to adapters the driver itself allocated. No public exploit code exists and EPSS is 0.17% (7th percentile), consistent with a niche, largely retired hardware platform.
NULL pointer dereference in the Linux kernel arm_mpam subsystem allows a local low-privileged attacker to crash the kernel by triggering a specific cleanup code path before initialization completes. The flaw exists in __destroy_component_cfg(), which unconditionally dereferences the embedded 'garbage' structure to free the configuration array even when mpam_disable() is called before __allocate_component_cfg() has ever run. No public exploit code exists and the EPSS score of 0.15% (5th percentile) confirms negligible real-world exploitation activity; vendor-supplied patches are available via upstream stable commits.
Signature type-binding bypass in @sigstore/core (npm) allows an attacker who controls the `payloadType` field of a DSSE envelope to substitute every ASCII character with a Unicode variant whose low byte matches, producing PAE bytes identical to a legitimate signature and causing verification to pass for a mismatched content type. All versions up to and including 3.2.0 are affected; a working proof-of-concept is included in the GitHub security advisory GHSA-jfc7-64v2-mr8c. The vulnerability is not confirmed in the CISA KEV catalog, but exploit code is publicly available, making exploitation straightforward for any attacker positioned to craft or relay DSSE envelopes.
Use-after-free in Envoy's HTTP OAuth2 filter (envoy.filters.http.oauth2) across versions 1.37.0-1.37.4 and 1.38.0-1.38.2 enables unauthenticated remote attackers to crash Envoy worker threads by racing a connection teardown against an in-flight async token exchange. When a downstream client disconnects while the filter is awaiting an async OAuth2 token response, the late AsyncClient callback invokes StreamDecoderFilterCallbacks on an already-freed object, producing undefined behavior and worker crashes. The reporter explicitly disclaims RCE - confirmed impact is availability loss (DoS); allocator-dependent memory corruption effects beyond crash are theoretically possible but not demonstrated. No public exploit or CISA KEV listing identified at time of analysis.
Use-After-Free crash in Envoy's ext_authz HTTP filter allows remote unauthenticated attackers to trigger a segmentation fault and process crash by rapidly establishing and tearing down downstream connections when per-route authorization overrides are configured. Affected are Envoy versions 1.36.0-1.36.8, 1.37.0-1.37.4, and 1.38.0-1.38.2. No public exploit code has been identified and this vulnerability is not listed in the CISA KEV catalog, though the race condition exploitability is constrained by requiring specific ext_authz per-route override configuration.
PROXY Protocol v2 TLV overflow in Envoy Proxy versions 1.34.0 through the pre-fix series allows an operator-level attacker on an adjacent network to smuggle extraneous bytes into upstream requests. Envoy's header generator writes TLV content exceeding the protocol-mandated 65535-byte maximum without adjusting the length field, creating a header where declared length and actual byte count diverge - a textbook CWE-130 length-parameter inconsistency. No public exploit code has been identified and this vulnerability does not appear in the CISA KEV catalog; risk is constrained by the high-privilege and adjacent-network prerequisites reflected in the CVSS 4.8 medium score.
Envoy proxy's External Processing (ext_proc) filter crashes with a use-after-free when an ext_proc gRPC server sends a single batched gRPC message containing multiple specially crafted ProcessingResponse objects. Affected versions span 1.34.0 through pre-fix releases across four maintained branches; fixed releases are 1.35.13, 1.36.9, 1.37.5, and 1.38.3. With CVSS PR:L, exploitation requires control of or compromise of the ext_proc gRPC server endpoint, limiting the realistic attacker population but making this a high-priority patch for any deployment running ext_proc. No public exploit code and no CISA KEV listing exist at time of analysis.
Certificate SAN validation in Envoy Proxy is bypassed when an attacker-controlled upstream serves a TLS certificate containing an embedded NUL byte in the dNSName Subject Alternative Name field. Envoy versions prior to 1.35.11, 1.36.7, 1.37.3, and 1.38.1 are affected, allowing an attacker who controls or can impersonate an upstream TLS endpoint to pass Envoy's configured SAN match check with a fraudulent certificate and intercept proxied traffic. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog.
The OAuth2 HTTP filter in Envoy Proxy prior to versions 1.35.11, 1.36.7, 1.37.3, and 1.38.1 implements AES-256-CBC encryption for the PKCE CodeVerifier cookie without any authentication tag, creating a classic padding oracle through differential HTTP responses on the /callback endpoint. An attacker who obtains the victim's encrypted CodeVerifier cookie and a stolen authorization code can recover the plaintext PKCE code_verifier in approximately 6,200 crafted requests (~100 seconds), then complete the OAuth token exchange to hijack the victim's access token. No public exploit code or CISA KEV listing has been identified at time of analysis; vendor-confirmed patches are available in all four current release branches.
Network annotation injection in KubeVirt's VirtualMachineInstance API allows authenticated tenants to attach launcher pods to arbitrary network namespaces by writing unsanitized JSON into the Multus CNI default-network annotation. Exploitation is gated on the ExternalNetResourceInjection Beta feature gate being explicitly enabled by a cluster-admin (off by default, first available in OpenShift Virtualization 4.21), which intentionally bypasses the NAD namespace lookup that would otherwise reject malformed annotation values. A successful attack permits cross-namespace network attachment and IP/MAC address impersonation on segments normally segregated from tenant workloads; no public exploit has been identified at time of analysis.
Cross-site scripting in jupyter/nbconvert versions 7.17.0 and earlier allows any user with notebook write access to inject and execute arbitrary JavaScript in the browsers of users who view HTML-exported notebooks. The flaw stems from the `data_mermaid` rendering block in `share/templates/lab/base.html.j2`, which outputs `text/vnd.mermaid` cell content directly into HTML without escaping, enabling tag breakout from the enclosing `<pre>` element. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog; however, the stored-XSS pattern means any user who views a maliciously crafted HTML export is at risk.
Use-after-free memory corruption in Google Chrome's Payments component on Android (prior to 149.0.7827.201) enables a local attacker with physical access to the device to trigger heap corruption, yielding high impact across confidentiality, integrity, and availability. The physical-access requirement (CVSS AV:P) substantially constrains the exploitable population to scenarios such as unattended or stolen devices. No public exploit identified at time of analysis, and no CISA KEV listing exists, indicating this has not been observed in active exploitation campaigns.
Lockdown mode in GitHub MCP Server HTTP deployments (versions 0.22.0-1.1.1) fails to isolate per-user GraphQL credentials due to a process-global singleton that retains the first authenticated user's client for the lifetime of the process. All subsequent users' IsSafeContent checks-the security gate that decides whether content from external contributors is sanitized before reaching the AI model-are evaluated under the first user's GitHub token and identity, producing wrong access control decisions for every user except the first. A publicly available proof-of-concept confirms the singleton pointer collision; no active exploitation (CISA KEV) has been confirmed at time of analysis.
Arbitrary filesystem write via zip-slip in K3s etcd snapshot restoration affects all K3s versions prior to 1.35.3+k3s1, 1.34.6+k3s1, and v1.33.10+k3s1. A maliciously crafted zip archive with path-traversal entries in member filenames (e.g., '../../etc/cron.d/evil') can overwrite arbitrary files on the host filesystem when an administrator performs a compressed etcd snapshot restore. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in CISA KEV.
Decompression bomb (zip bomb) in Trivy's Helm chart scanner causes denial of service via OOM kill when processing a crafted .tgz archive. Affected versions prior to 0.71.0 use io.ReadAll without any size cap on tar entries inside Helm chart archives, allowing a small compressed file to expand to gigabytes in memory. An attacker who can place a malicious archive in any path that Trivy is directed to scan can reliably crash the Trivy process. No public exploit or CISA KEV listing at time of analysis.
Stack out-of-bounds write in Vim's spell suggestion engine allows a crafted `.spl`/`.sug` file pair to corrupt the call stack and crash the editor when spell suggestions are invoked. All Vim releases prior to 9.2.0653 are affected, with the flaw in `tree_count_words()` and `sug_filltree()` inside `src/spellfile.c`, where three fixed-size MAXWLEN-element stack arrays are indexed by an unbounded depth counter. No public exploit identified at time of analysis (CVSS 4.0 E:U), though the upstream fix commit includes a detailed proof-of-concept test case reproducing the out-of-bounds write.
Out-of-bounds read in Vim's text property subsystem allows a crafted undo file to crash the editor and potentially disclose adjacent heap memory. All Vim releases prior to 9.2.0670 are affected; the vulnerability is reachable whenever a user loads a malicious undo file with the persistent-undo feature compiled in and enabled. No public exploit or CISA KEV listing exists at time of analysis, placing real-world risk in the low-to-medium range despite the local crash severity.
Vim crashes with an out-of-bounds read when opening a maliciously crafted file encrypted with the xchacha20poly1305 cipher (VimCrypt~04! or VimCrypt~05!) whose body is shorter than a libsodium secretstream header. An unsigned integer underflow in the length calculation inside crypt_sodium_buffer_decode() causes the subsequent crypto_secretstream_xchacha20poly1305_pull() call to read far past the end of the input buffer, resulting in a denial-of-service crash. No public exploit code and no CISA KEV listing have been identified at time of analysis; however, the fix commit includes a concrete reproducer test case confirming the crash path.
Out-of-bounds heap read in Vim 9.2.0320-9.2.0678 allows a crafted undo or swap file to disclose adjacent heap memory or crash the editor when a victim opens it. The flaw resides in the text property (textprop) subsystem: a maliciously encoded virtual-text property with an oversized tp_text_offset is converted directly to a heap pointer without bounds validation in both the undo-restoration path (um_goto_line) and the display path (get_text_props). The CVSS 4.0 score of 6.8 (Medium) reflects local-only exploitation requiring active user interaction; no public exploit is confirmed at time of analysis, though the fix commit includes detailed test scaffolding constituting a near-complete attack blueprint.
Stack out-of-bounds write in Vim's spell_soundfold_sofo() function (src/spell.c) allows local exploitation when a SOFO-based spell language is active and the soundfold path is triggered with a word exceeding MAXWLEN (254) bytes, corrupting the call frame and crashing the editor with theoretical code execution potential. All Vim releases prior to 9.2.0698 carrying the single-byte SOFO branch are affected; the vendor has released a confirmed fix at v9.2.0698. No public exploit code exists and no CISA KEV listing is present; the CVSS 4.0 E:U supplemental metric confirms exploitation remains unproven at time of analysis.
Out-of-bounds read in Nokogiri's `NodeSet#[]` method (all versions before 1.19.4) enables an attacker who can supply a large negative integer index to bypass a 32-bit-truncated bounds check and trigger a read outside the node set's allocated memory. On CRuby (MRI Ruby), this typically crashes the process - denial of service - with ancillary potential for heap memory disclosure; on JRuby, the runtime's managed memory prevents memory corruption, but an incorrect node is silently returned. No public exploit or active exploitation has been identified; exploitation requires the target application to forward attacker-controlled integers directly to the `NodeSet#[]` or `#slice` API.
DNS response manipulation in PowerDNS Recursor 5.4.x allows network-positioned attackers to inject false DNS records through insufficient validation of answers received from authoritative servers, resulting in low-integrity violations in name resolution for downstream clients. The 5.4.x branch received hardening patches per PowerDNS advisory 2026-08 to close this validation gap. No public exploit has been identified at time of analysis and the vulnerability is not listed in the CISA KEV catalog.
Denial of service in PowerDNS dnsdist's DNS over HTTP/3 (DoH3) handler allows unauthenticated remote attackers to exhaust server memory by opening high volumes of concurrent QUIC streams carrying crafted queries. Each crafted query triggers an exception that defers buffer deallocation until QUIC connection teardown rather than releasing memory promptly, enabling cumulative memory exhaustion across simultaneous streams. No public exploit code or active exploitation has been identified at time of analysis, and the attack surface is limited to instances with DoH3 explicitly enabled.
dnsdist's SetMacAddrAction handler exposes operators to uninitialized memory leakage in DNS responses and potential service crashes when the action is configured in the ruleset. The flaw is reachable over the network without authentication (AV:N/PR:N), but the high attack complexity (AC:H) constrains real-world impact to deployments that have explicitly enabled SetMacAddrAction - a non-default configuration. No public exploit code exists and no CISA KEV listing is present at time of analysis; the PowerDNS security team (Open-Xchange) reported this internally, suggesting responsible disclosure rather than observed active exploitation.
TCP connection resource leak in PowerDNS dnsdist allows remote unauthenticated attackers to exhaust backend TCP connection capacity by sending IXFR (Incremental Zone Transfer) queries. Affected dnsdist instances fail to promptly release outgoing TCP connections to backends after IXFR processing, leaving them open until OS-level timeout; under sustained query volume this can exhaust the backend's concurrent connection limit or the dnsdist process's file descriptor table. No public exploit has been identified at time of analysis, and no CISA KEV listing exists, but the AV:N/AC:L/PR:N attack surface makes the trigger trivially automatable from the internet.
DNSSEC validation in PowerDNS Recursor can be silently undermined by network-positioned attackers who spoof DNS replies to trick the Recursor into permanently marking a legitimate authoritative server's IP as EDNS-incapable. Once that state is poisoned, the Recursor is unable to perform proper DNSSEC validation for any zones served by that authoritative server, effectively disabling a critical security layer for downstream clients. No active exploitation has been confirmed by CISA KEV, and no public proof-of-concept has been identified at time of analysis; however, the attack complexity is rated High due to the spoofing prerequisite.
ZONEMD validation bypass in PowerDNS Recursor allows a cryptographically invalid DNS zone to be accepted as legitimate when ZoneToCache is configured with ZONEMD validation enabled. This undermines the integrity guarantee that ZONEMD is designed to provide - an attacker capable of serving a crafted zone to the resolver (e.g., via a rogue authoritative source or network interception) can cause the recursor to cache and serve tampered DNS data. No public exploit code has been identified and exploitation is not confirmed in CISA KEV; real-world risk is constrained by the non-default configuration prerequisite.
Cross-origin data leakage via Chrome's Autofill implementation on Android allows a remote attacker with prior renderer process compromise to extract sensitive data across origin boundaries using a crafted HTML page. Affected versions are all Chrome for Android releases prior to 150.0.7871.47. No public exploit code has been identified and the EPSS score of 0.21% (11th percentile) reflects low observed exploitation activity; however, the cross-origin data exposure (CWE-346) combined with Autofill access creates meaningful privacy risk for users on vulnerable Android builds. Google has released a patch in the stable channel update.
Navigation restriction bypass in Google Chrome's Passwords subsystem allows a remote attacker to circumvent intended access controls by delivering a crafted HTML page to a user. All Chrome desktop versions prior to 150.0.7871.47 are affected, with integrity rated High by CVSS due to the ability to manipulate navigation behavior. No public exploit exists at time of analysis and EPSS sits at 0.22% (13th percentile), indicating low near-term exploitation probability despite Chrome's enormous global footprint and Chromium's internal 'High' severity classification.
Cross-origin data leakage in Google Chrome for Android prior to 150.0.7871.47 enables remote attackers to read data from other origins via a crafted HTML page exploiting insufficient validation in the File Input component. The flaw is Android-platform-specific, requiring only that a victim user navigate to an attacker-controlled page. No active exploitation has been confirmed (CISA KEV negative, SSVC exploitation: none), and EPSS at 0.21% (11th percentile) reflects low current exploitation interest, though the unauthenticated, low-complexity attack vector keeps it a relevant patching priority for Android Chrome deployments.
Universal Cross-Site Scripting (UXSS) in Google Chrome for iOS prior to 150.0.7871.47 enables a remote attacker to inject arbitrary scripts or HTML across origins by exploiting insufficient input validation triggered through specific user UI gestures on a crafted page. The scope change (S:C in CVSS) is the critical dimension here - successful exploitation bypasses the same-origin policy, potentially granting the attacker script execution in the context of arbitrary origins within the browser session. No public exploit code has been identified and EPSS sits at just 0.20% (11th percentile), indicating low observed exploitation activity despite the high-impact class of vulnerability.
Memory disclosure in Google Chrome for Linux (prior to 150.0.7871.47) allows a remote attacker to read sensitive contents from the browser's process memory by directing a user to a crafted HTML page. The flaw stems from an inappropriate implementation in Chrome's input handling subsystem, and is platform-specific - only Linux builds are affected. No public exploit code or CISA KEV listing has been identified at the time of analysis, but the network-reachable, zero-authentication attack path and high confidentiality impact make patching a near-term priority for Linux desktop environments.
Sensitive process memory exposure in Google Chrome for iOS prior to 150.0.7871.47 enables a physically present attacker to read potentially sensitive data from the browser's memory without authentication. The root cause is insufficient data validation (CWE-20) within the iOS-specific Chrome implementation. No public exploit code or CISA KEV listing has been identified at time of analysis, and the physical access requirement (CVSS AV:P) significantly constrains real-world attacker opportunity.
Navigation restriction bypass in Google Chrome for iOS (prior to 150.0.7871.47) permits a remote, unauthenticated attacker to circumvent client-side policy controls by delivering a crafted HTML page to a victim user. The flaw originates from insufficient policy enforcement (CWE-602), where navigation security logic enforced within the browser client can be subverted through malicious web content. No active exploitation has been identified - CISA KEV is not listed, SSVC confirms Exploitation: none, and the EPSS score of 0.22% (13th percentile) reflects minimal real-world exploitation activity at time of analysis.
Cross-origin data leakage in Google Chrome's SVG rendering subsystem (versions prior to 150.0.7871.47) allows remote attackers to exfiltrate sensitive information from other origins by directing victims to a crafted HTML page. The root cause is insufficient policy enforcement in SVG handling, classified as CWE-346 (Origin Validation Error), effectively bypassing Same-Origin Policy protections. No public exploit has been identified at time of analysis, and EPSS places exploitation probability at 0.21% (11th percentile), indicating low near-term exploitation likelihood despite a CVSS confidentiality impact rated High.
Side-channel information leakage via Chrome's Scroll implementation exposes cross-origin data to remote attackers who can lure a victim to a crafted HTML page. All Chrome versions prior to 150.0.7871.47 are affected; exploitation requires user interaction (visiting the attacker-controlled page) but no authentication. No public exploit identified at time of analysis and EPSS sits at the 11th percentile, though the confidentiality impact is rated High by NVD given the potential to read data from foreign origins.
ImageMagick before 7.1.2-24 fails to correctly enforce path restrictions configured in its security policy (policy.xml), enabling file creation and truncation outside designated filesystem boundaries. In sandboxed image conversion services - a common deployment pattern for web applications accepting user-submitted images - this incorrect policy check can be exploited by attackers who submit crafted images to write arbitrary files accessible to the conversion process. No public exploit has been identified at time of analysis and this is not listed in CISA KEV, but the wide use of ImageMagick in web application stacks makes the affected population substantial.
Nonce reuse in ImageMagick's AES-CTR cipher implementation exposes encrypted image plaintext to recovery attacks. The PasskeyEncipherImage method in ImageMagick before 7.1.2-22 reuses nonces when performing AES in Counter mode, violating the fundamental security requirement that a nonce be used exactly once per key. A network-accessible attacker who can collect multiple ciphertexts produced with the same passkey and nonce can XOR the outputs to cancel the shared keystream and partially or fully recover encrypted image content. No active exploitation has been identified (not in CISA KEV), and no public exploit code is known at time of analysis.
Memory exhaustion via denial of service in ImageMagick before 7.1.2-19 allows remote attackers to trigger a persistent memory leak in the PNG encoder by supplying crafted MNG image input that induces an encoder failure condition. Affected deployments include any service that accepts user-supplied images and processes them through ImageMagick's MNG write path. No public exploit code or active exploitation has been identified at time of analysis, but the network-reachable attack surface in image-processing pipelines makes this a realistic DoS risk against unpatched instances.
Division by zero in ImageMagick's binomial kernel processing path, affecting all versions before 7.1.2-22, allows an attacker to crash the application by supplying a crafted image with an oversized binomial kernel value. The integer overflow corrupts the kernel size calculation, and a subsequent division by the resulting zero value causes an unhandled exception and process termination. No public exploit code has been identified and the vulnerability is not listed in CISA KEV; however, automated image-processing pipelines that accept untrusted uploads are at elevated practical risk despite the local-vector CVSS rating.
Heap buffer out-of-bounds read in ImageMagick before 7.1.2-19 via off-by-one error in morphology validation allows local attackers with user interaction to trigger denial of service. The vulnerability stems from incorrect morphology parameters causing single pixel memory access violations within heap buffers, potentially leading to application crashes or information disclosure through controlled reads of adjacent heap memory.
### Summary Responsibly disclosed by @NSEcho. Rated medium severity (CVSS 4.9), this vulnerability is remotely exploitable, low attack complexity. This Uncontrolled Resource Consumption vulnerability could allow attackers to cause denial of service by exhausting system resources.
### Impact Shovel and Federation plugins perform URI obfuscation in their worker (link) state. Rated medium severity (CVSS 5.5), this vulnerability is low attack complexity.
Path traversal in Rancher Fleet's ImageScan subsystem (CWE-23) allows authenticated remote attackers to escape intended directory boundaries and trigger denial of service across four active release branches (0.12.x through 0.15.x). The flaw appears to have been introduced in the 0.12.0 series and persisted undetected through at least sixteen subsequent patch releases, indicating the ImageScan component lacked adequate path sanitization for an extended period. No public exploit code or active exploitation has been identified; the CVSS 4.0 vector (PR:L) confirms exploitation requires authenticated access, materially limiting opportunistic attack surface.
Privilege persistence in SUSE Rancher allows project users to retain Pod Security Admission (PSA) permissions even after an administrator revokes those permissions from a RoleTemplate, due to a missing cleanup step in the legacy Project Role Template Binding (PRTB) reconciler. Affected versions are Rancher 2.13.0 through 2.13.7 and 2.14.0 through 2.14.3. An attacker with a pre-existing role assignment can continue to bypass PSA policies enforced at the project level, defeating administrative intent and potentially deploying workloads that violate the cluster's security posture. No public exploit has been identified at time of analysis, and the flaw is not listed in CISA KEV.
Denial of service in fzf's --listen mode HTTP server arises from quadratic-time (O(n²)) body processing caused by repeated string concatenation when reading POST request payloads. All fzf releases prior to 0.73.1 are affected when the --listen flag is explicitly activated to expose the embedded HTTP server. An attacker with access to the listening socket can send a single crafted POST request with many small body segments to monopolize the single-threaded server, blocking all other clients indefinitely. No public exploit identified at time of analysis; vendor-released patch is available in version 0.73.1.
Integer overflow in fzf's FuzzyMatchV2 function causes a non-recoverable process crash (denial of service) when input lines reach approximately 2,200,000 bytes and a search pattern of ~999 bytes is active. The Go runtime detects the resulting invalid slice bounds and immediately terminates the process with a panic - there is no recovery path and no possibility of memory corruption or code execution. No public exploit code has been identified and this CVE is not listed in CISA KEV; the CVSS 4.0 score of 5.6 (AT:P, UI:A) reflects that exploitation requires atypical input volumes and active user or script interaction.
Super-linear algorithmic complexity in the decode-uri-component npm package (versions through 0.4.1) allows unauthenticated network attackers to cause severe CPU exhaustion and block the Node.js event loop by submitting crafted strings with large numbers of percent-encoded tokens. The decode() function splits input on '%' and iteratively invokes decodeComponents() in a pattern producing quadratic or worse time growth - 1,400 '%ab' tokens consume approximately 33 seconds of CPU, rendering single-threaded JavaScript runtimes effectively unresponsive. No public exploit is confirmed at time of analysis (CVSS E:U), but the deterministic algorithmic nature makes payload construction trivial for any attacker with a reachable input channel.
Insufficient data exists to characterize CVE-2026-13606. The sole intelligence signal is a vendor report attributed to Ubuntu, indicating this vulnerability affects a package or component within the Ubuntu ecosystem. No description, CVSS score, vector, CWE classification, or additional references are available - a complete technical characterization is not possible from the provided data.
Improper Authorization (CWE-285) in Apache Tomcat's default servlet allows HTTP method-based and method-omission security constraints to be silently bypassed across all major supported Tomcat branches from 7.0.x through 11.0.x. An attacker can perform HTTP operations - such as PUT or DELETE - that the web.xml security constraint configuration was intended to restrict, potentially enabling unauthorized file upload, modification, or deletion on the default servlet's served content. No active exploitation has been confirmed (not in CISA KEV) and no CVSS score has been published at time of analysis, but the broad version range and ubiquitous deployment footprint of Tomcat make prompt patching a priority.
Replay attack vulnerability in Apache Tomcat's cluster EncryptionInterceptor allows a network-adjacent attacker to retransmit previously captured encrypted inter-node cluster messages, causing receiving nodes to accept and process them as legitimate - potentially corrupting distributed session state or triggering unintended cluster actions. All major supported branches are affected: 11.0.0-M1 through 11.0.22, 10.1.0-M1 through 10.1.55, 9.0.13 through 9.0.18, plus end-of-life branches 8.5.38-8.5.100 and 7.0.100-7.0.109. No public exploit or CISA KEV listing has been identified at time of analysis; however, the breadth of affected versions across all active and legacy branches elevates organizational exposure, particularly for environments running EOL 8.5.x or 7.x with no available patch.
Reflected XSS in Apache Tomcat's bundled 'number guess' example application exposes users of that demo page to script injection across all major Tomcat release lines from 7.0 through 11.0. The flaw resides in a sample JSP/servlet, not the core Tomcat runtime, meaning exploitation depends entirely on the example application being deployed and accessible - a configuration that violates standard production hardening guidance. No public exploit code or active exploitation has been identified at time of analysis; no CVSS vector was assigned by the reporter.
Memory corruption (CWE-119) in Apple Safari's web content processing engine causes an unexpected application crash when rendering maliciously crafted web content. Affected across Safari, iOS 26, iPadOS 26, and macOS Tahoe - all versions prior to 26.5.2. An unauthenticated remote attacker can trigger a denial-of-service against any user who visits a hostile page, though no code execution is indicated by current data. No public exploit code and no CISA KEV listing are identified at time of analysis; exploitation probability (EPSS) was not provided in source data.
Server-side request forgery in Pinpoint APM through version 3.1.0 allows authenticated users with low privileges to register internal or cloud-metadata URLs as alarm webhook targets due to missing URL validation on the webhook registration endpoint. By deliberately triggering alarm threshold breaches, an attacker can coerce the Pinpoint server to issue outbound POST requests to internal hosts, cloud metadata services (such as AWS IMDS at 169.254.169.254), or other non-routable internal infrastructure. No public exploit code or CISA KEV listing has been identified at time of analysis, though the CVSS 4.0 subsequent-system confidentiality impact is rated High, reflecting the potential to enumerate internal services or harvest cloud credentials.
Path traversal in spice-vdagent enables a malicious or compromised SPICE host to write arbitrary files to any location on the guest operating system during file transfer operations. Affecting Red Hat Enterprise Linux versions 6 through 10, the flaw stems from unsanitized filenames supplied by the SPICE host before use by the vdagent process, allowing writes at the privilege level of the logged-in guest user. No public exploit is identified at time of analysis, and exploitation requires prior control of the SPICE host - a high-privilege precondition that substantially constrains realistic risk despite the high integrity impact.
Integer overflow in spice-vdagent's message parsing allows a malicious or compromised SPICE host to crash the guest VM's vdagent daemon by delivering a specially crafted message that triggers a downstream heap buffer overflow. Affected deployments span spice-vdagent as packaged in Red Hat Enterprise Linux 6 through 10, where the daemon runs inside guests handling host-to-guest SPICE protocol communications. No public exploit code or CISA KEV listing exists at time of analysis; exploitation is structurally constrained by the requirement that the attacking SPICE host itself be untrusted or already compromised prior to the attack.
Path traversal in ANTLR4 up to 4.13.2 exposes arbitrary file read via the tokenVocab grammar option handler. The vulnerable function getImportedVocabFile in TokenVocabParser.java fails to sanitize user-controlled grammar option values, allowing an attacker to supply a crafted .g4 grammar file that causes the tool to read files outside the intended working directory. A publicly available proof-of-concept exploit exists (GitHub issue reference), and no vendor patch has been released - the vendor did not respond to disclosure. No active exploitation is confirmed in CISA KEV, but the CVSS 4.0 E:P flag confirms exploit code is public.
Code injection in ANTLR4's Grammar Action Block Handler allows remote attackers to execute arbitrary code by supplying a crafted grammar file processed by the tool. All versions up to and including 4.13.2 are affected via the OutputFile.java code generation pathway. A public proof-of-concept exploit exists on GitHub (no KEV listing), though the vendor has not acknowledged or patched the issue, leaving users without an official fix.
Out-of-bounds reads and a process crash in Nmap through 7.99 can be triggered remotely by any host that is being scanned, or by an on-path attacker, via a crafted IPv6 packet with a truncated extension header. The root cause is an integer underflow in the `ipv6_get_data_primitive` function (libnetutil/netutil.cc) where an insufficiently strict bounds check allows the extension-header walk pointer to advance past the captured packet buffer, causing the remaining-length subtraction to wraparound to a very large value. A publicly available exploit proof-of-concept exists; this vulnerability is not currently listed in the CISA KEV catalog.
HTTP request/response smuggling in nghttpx (the reverse proxy component of nghttp2 through 1.69.0) allows unauthenticated remote attackers to poison shared backend keep-alive connections by crafting an HTTP/1.1 Upgrade request that simultaneously carries a Content-Length header and body. When nghttpx forwards this ambiguous message to a backend and re-adds Connection and Upgrade headers while passing Content-Length verbatim, a backend that resolves the parsing ambiguity in the attacker's favor treats the body as a separate, attacker-controlled HTTP request - enabling cross-client response-queue poisoning. A publicly available proof-of-concept exploit exists; no CISA KEV listing at time of analysis.
NULL pointer dereference in the Linux kernel MANA (Microsoft Azure Network Adapter) driver crashes the host kernel when initializing Virtual Function (VF) devices in VFIO passthrough or nested KVM environments where `pdev->slot` is unpopulated. Systems running affected kernel versions in Azure-style or nested virtualization deployments face deterministic kernel panics and total availability loss on the host. No public exploit code has been identified, and EPSS probability is 0.16% (5th percentile), indicating low opportunistic exploitation risk - though in targeted Azure or nested KVM environments, the crash is straightforwardly reproducible.
Deadlock in the Linux kernel's DSA (Distributed Switch Architecture) subsystem allows a local low-privileged user to hang the kernel network stack by running ethtool against a DSA-configured conduit interface. The DSA subsystem replaces the conduit device's ethtool_ops with aggregating wrappers, but these wrappers redundantly invoke netdev_lock_ops() on a lock already held by the kernel ethtool infrastructure, creating a classic double-lock deadlock. No public exploit exists and EPSS sits at 0.15% (5th percentile), though the issue is trivially reproducible with standard system tooling on affected configurations.
Unbounded NAPI busy-poll loop in the Linux kernel's io_uring subsystem allows a low-privileged local process to spin the kernel indefinitely when no I/O events arrive, triggering soft lockup watchdog complaints and degrading system availability. Affected kernels are those at or after commit 8d0c12a80cdeb80d5e0510e96d38fe551ed8e9b5 (introducing NAPI support in io_uring, Linux 6.9) through the patched stable releases 6.18.33, 7.0.10, and 7.1. No public exploit exists and this is not listed in CISA KEV; EPSS sits at 0.15% (5th percentile), reflecting very low exploitation interest.
Improper input validation in the Linux kernel's nilfs2 filesystem driver allows a local low-privileged user to trigger a kernel WARN_ON and cause a denial of service by submitting a crafted ioctl request with bd_oblocknr set to zero. The zero value collides with the -ENOENT sentinel path in nilfs_ioctl_mark_blocks_dirty(), bypassing the dead block detection logic and passing an invalid block reference into nilfs_bmap_mark(). No active exploitation has been identified (EPSS 0.17%, not in CISA KEV), and patches are available across all actively maintained Linux stable branches.
Spurious WARN_ON_ONCE calls in the Linux kernel's blk-wbt subsystem trigger unnecessary kernel warnings and stack traces under expected failure conditions - memory pressure or a pre-registered writeback throttle. Local low-privileged users can provoke this via ioctl(BLKPG) during MTD partition creation under memory-constrained conditions, generating noise in kernel logs and, on panic_on_warn-enabled kernels, potentially causing a system crash. No public exploit exists and the EPSS score of 0.14% (4th percentile) places this firmly in the low-priority tier; it has not been added to CISA KEV.
NULL pointer dereference in the Linux kernel's mt76/mt7925 WiFi driver crashes the kernel when the 'sta' pointer is dereferenced before a NULL check in mt7925_tx_check_aggr(). Systems with MediaTek mt7925 WiFi hardware running affected kernel versions are exposed to a local denial-of-service via kernel panic. No public exploit exists and EPSS is at the 5th percentile (0.16%), indicating no observed widespread exploitation; patches have been applied across multiple stable kernel branches.
Firmware crash in the Linux kernel's mt76/mt7921 WiFi driver allows a local low-privileged user to cause denial of service by triggering station Association ID (AID) values above the firmware's undocumented limit of 20. Affected hardware includes Mediatek mt7922 chipsets operating in AP mode (IFTYPE_AP). Exploitation requires running a non-standard hostapd configuration that allocates AIDs starting at 65 or higher - stock hostapd is not affected. No public exploit has been identified and EPSS is 0.16% (5th percentile), indicating very low exploitation likelihood.
NULL pointer dereference in the Linux kernel's AMD RAS subsystem crashes systems with AMD GPU hardware. The function `ras_core_ras_interrupt_detected()` in `drm/amd/ras` fails to validate the `ras_core` pointer before dereferencing `ras_core->dev` in an error path, enabling a local low-privilege user to trigger a kernel panic and denial of service. No public exploit code exists and the vulnerability is not listed in CISA KEV; EPSS at 0.14% (4th percentile) confirms negligible current exploitation probability. Upstream fix commits are available in stable kernel trees.
NULL pointer dereference in the Linux kernel's drm/amd/ras subsystem allows a local low-privileged user to crash the kernel on systems with AMD GPUs. The flaw in ras_core_get_utc_second_timestamp() occurs when a NULL ras_core pointer passes a conditional check but is then dereferenced in the subsequent error-printing path, producing a kernel panic. No public exploit exists and EPSS sits at 0.15% (4th percentile), placing this firmly as a low-urgency local denial-of-service hardening issue despite the Availability:High CVSS rating.
Denial of service in the Linux kernel's padata parallel data processing subsystem results from a misplaced CPU hotplug callback that returns an error in a section where error returns are not permitted. Specifically, padata_cpu_dead() is registered below CPUHP_TEARDOWN_CPU, a hotplug state phase where callbacks are not allowed to fail; when the callback returns non-zero, the kernel emits a 'DEAD callback error' warning that can destabilize the system. A local attacker with low-privilege access on a multi-CPU system using padata (e.g., for parallel IPsec or crypto processing) can trigger this condition during CPU offline transitions, resulting in high availability impact. No public exploit is identified and EPSS is 0.16%, indicating low exploitation probability; patches are available across multiple stable kernel branches.
Integer overflow on the RISC-V IOMMU invalidation path in the Linux kernel triggers an infinite loop, causing a local denial of service. A low-privileged local user on an affected RISC-V system with IOMMU hardware can hang the kernel by inducing an invalidation where gather->end equals ULONG_MAX, exploiting the sign-extended page table range supported by RISC-V. EPSS is 0.15% (5th percentile), no active exploitation has been identified, and no public exploit code is known.
Uninitialized memory read in the Linux kernel FUSE filesystem driver causes local denial-of-service conditions. Specifically, `fuse_dentry_revalidate()` in `fs/fuse/dir.c` reads the `->d_time` field of a dentry that was allocated by `__d_alloc()` without initialization, detectable by KMSAN as a use-before-initialize defect. A local authenticated user with access to a FUSE-mounted filesystem can trigger a kernel crash by opening a file through the affected `lookup_open` → `d_revalidate` code path. No public exploit exists and EPSS is 0.15%, indicating minimal real-world exploitation pressure at this time.
Kernel page fault in the Linux kernel's Tegra234 Control Backbone (CBB) driver crashes affected NVIDIA Tegra234-based systems when a cross-fabric error interrupt triggers incorrect register address resolution. The `tegra234_cbb_get_tmo_slv()` function uses the receiving fabric's base address (`cbb->regs`) with register offsets derived from a different fabric's target map, producing an invalid virtual address dereference (`ffff80000954cc00`) and a kernel panic. Exploitation requires local low-privilege access to Tegra234 hardware; no public exploit or POC exists, EPSS sits at the 5th percentile (0.15%), and the vulnerability is not in CISA KEV.
Null pointer dereference in the Linux kernel's pinctrl subsystem allows a local low-privileged attacker to crash the kernel, causing a denial of service. The flaw exists in pinconf_generic_parse_dt_pinmux(), which fails to validate that a 'pinmux' Device Tree property is non-empty before use - an empty property causes the allocator to return an invalid non-NULL pointer, and subsequent memory access triggers a kernel panic. No public exploit code exists and EPSS is 0.15% (5th percentile), reflecting low real-world exploitation probability; exploitation is further constrained to systems with specific pinctrl hardware and Device Tree configurations.
Off-by-one array overflow in the Linux kernel's hvc_iucv driver (IBM z/VM IUCV HVC terminal subsystem) can trigger an out-of-bounds write to hvc_iucv_table[8] when hvc_iucv_devices is set to the maximum value of 8, crashing the kernel. Exploitation requires local access with low privileges on IBM System z systems running z/VM with hvc_iucv configured. No public exploit identified at time of analysis; EPSS is 0.18% (7th percentile), consistent with a niche mainframe-only driver affecting a very limited deployment population.
NULL pointer dereference in the Linux kernel's ps883x USB Type-C retimer driver causes a kernel Oops during device unbind, resulting in a denial of service. The ps883x_retimer_remove() function calls i2c_get_clientdata() to retrieve driver-private state, but i2c_set_clientdata() was never called during probe, leaving the pointer NULL; dereference at offset 0x20 triggers a fatal translation fault. Systems with ps883x retimer hardware running affected kernel versions are vulnerable when a privileged local user unbinds the driver via sysfs. No active exploitation is confirmed (no CISA KEV listing), and the EPSS score of 0.17% (6th percentile) reflects negligible real-world exploitation probability.
Soft CPU lockup in the Linux kernel SCSI generic (sg) driver allows a local user to trigger a denial of service by writing an out-of-range value directly to the def_reserved_size sysfs module parameter, bypassing the validation enforced by sg_proc_write_dressz. Setting def_reserved_size to -1 via /sys/module/sg/parameters/def_reserved_size causes sg_build_reserve to enter a non-terminating allocation loop on the next open() of any /dev/sgX device, hanging the affected CPU core for 26+ seconds until the kernel watchdog fires. No public exploit code has been identified and EPSS probability is very low at 0.18%; patches are confirmed available across all active stable kernel branches.
NULL pointer dereference in the Linux kernel's EIP93 hardware crypto driver triggers a kernel panic during AEAD crypto operations on affected embedded systems. The root cause is a self-contradictory algorithm lookup mask in eip93_hmac_setkey() - CRYPTO_ALG_ASYNC is passed as an exclusion mask, yet EIP93 algorithms are inherently async, so the lookup always returns -ENOENT. This leaves the AEAD security association record with zeroed digest fields; any subsequent crypto operation then dereferences a NULL context pointer, crashing the kernel. No public exploit is identified at time of analysis, and EPSS is very low (0.17%, 6th percentile), reflecting the narrow hardware-specific attack surface.
Null pointer dereference in the Amlogic T7 SoC reset driver crashes the Linux kernel when uninitialized reset operation callbacks are invoked, causing a full system denial of service. Local low-privilege users on systems physically equipped with an Amlogic T7 SoC are affected across kernel versions from the introduction commit through the patched releases 6.18.33, 7.0.10, and 7.1. No public exploit has been identified and EPSS of 0.17% (6th percentile) reflects negligible real-world exploitation interest; this is a hardware-specific stability fix rather than an actively exploited security threat.
NULL pointer dereference in the Linux kernel's Airoha QDMA network driver causes a kernel panic (denial of service) when queue entry list allocation fails during TX queue initialization. Systems running Airoha-based network hardware on affected kernel versions - including Linux 6.19 - are vulnerable when the driver's cleanup path dereferences an uninitialized descriptor array pointer. No public exploit exists and EPSS sits at 0.17% (6th percentile), indicating minimal real-world exploitation activity; no CISA KEV listing.
NULL pointer dereference in the Linux kernel's airoha QDMA network driver crashes the kernel when RX queue initialization fails mid-way through allocation. The flaw exists in airoha_qdma_init_rx_queue() where early ndesc initialization causes the cleanup routine to call netif_napi_del() on NAPI structures never registered via netif_napi_add(), dereferencing a null pointer and causing a local denial of service. No public exploit identified and EPSS at 0.17% (6th percentile) reflects minimal real-world exploitation probability; impact is constrained to systems running Airoha QDMA-based network hardware.
Kernel panic via NULL pointer dereference in the Linux kernel MANA (Microsoft Azure Network Adapter) driver allows a local low-privileged user to crash the host when a power management resume failure leaves the driver in a torn-down state. The mana_remove() function is invoked twice - first during a failed mana_probe() recovery path that nullifies gdma_context and driver_data, and again when the driver is subsequently unbound - dereferencing a NULL gc pointer. No public exploit exists and EPSS is 0.17% (6th percentile), indicating minimal real-world exploitation activity; however, the impact on Azure Linux VM availability is complete (kernel panic).
Improper locking in the Linux kernel's amdgpu DRM driver allows a local low-privileged user to trigger a kernel deadlock or system crash via the AMDGPU_INFO_READ_MMR_REG ioctl. Three distinct concurrency defects exist: inverted lock ordering between the reset semaphore and mm_lock (permitting copy_to_user() under lock), memory allocation while holding the reset semaphore, and use of down_read_trylock() where a blocking wait is required. No public exploit code has been identified and EPSS is 0.17%, placing this firmly in opportunistic-rather-than-targeted risk territory.
Kernel crash via NULL pointer dereference in the ALSA HDA Conexant audio driver affects Linux kernel stable branches from 5.15 through 7.0, exploitable by a local low-privileged user on systems with Conexant HDA audio hardware. The cx_probe() function ignores the error pointer returned by snd_hda_jack_detect_enable_callback() on memory allocation failure, leaving the jack detection callback unregistered and an internal structure uninitialized; a subsequent audio jack event then triggers a kernel panic. No public exploit exists and no CISA KEV listing is present; EPSS is 0.17% (7th percentile), confirming very low real-world exploitation probability.
NULL pointer dereference in the Linux kernel ice driver's ice_reset_all_vfs() function crashes the host kernel when a Virtual Function VSI rebuild fails during concurrent NVM firmware updates on Intel E800-series NICs. Systems with SR-IOV enabled on Intel E800-series adapters running kernel versions from 5.8 through the patched stable releases are affected. An attacker or privileged local process can trigger a kernel panic by initiating a VF reset while nvmupdate64e has placed Admin Queue commands in a transitional failure state, resulting in complete host denial of service. No public exploit exists and EPSS is negligible at 0.17% (7th percentile); CISA KEV listing is absent.
Memory corruption in the arm64 early kernel mapping on Linux allows a local low-privileged user to cause a kernel panic through an off-by-one page-table allocation error introduced since commit 5973a62efa34. Affected systems running 4K page granule on arm64 hardware may have the [data, end) segment overflow into the guard gap page before early_init_stack, corrupting critical early initialization memory. No public exploit exists and EPSS is 0.17% (6th percentile), reflecting this as primarily a kernel stability and reliability concern for arm64 deployments rather than an actively weaponized attack vector.
Silent audit trail corruption in the Linux kernel's capability auditing subsystem allows a local attacker who manipulates inheritable capabilities as a precursor to privilege escalation to have that manipulation go unrecorded in CAPSET audit logs since 2008. The `__audit_log_capset()` function records `cap_effective` into the `cap_pi` (inheritable) field due to a copy-paste error, meaning compliance and forensic systems receive falsified capability state that masks the inheritable-cap modification step of a privilege escalation chain. No public exploit has been identified at time of analysis, and the EPSS score of 0.18% (7th percentile) reflects that exploitation as a standalone attack is not observed in the wild.
Boot-time kernel crash in Linux 6.16+ on AMD IOMMU-equipped systems causes a General Protection Fault when a PCI device whose Bus:Device.Function address is absent from the ACPI IVRS table is encountered during IOMMU initialization. The root cause is a missing bounds check in __rlookup_amd_iommu() that was latent until commit e874c666b15b changed the rlookup_table allocation from a zeroed page-order block (which returned NULL on overrun) to a tight kvcalloc(), causing adjacent slab contents to be dereferenced as a valid struct amd_iommu pointer. The result is a non-recoverable GPF at boot time, confirmed in production on Google Compute Engine ct6e VMs; no public exploit code and no CISA KEV listing exist at time of analysis.
Stack access fault in the x86/kexec purgatory handoff crashes non-kjump kexec operations on affected Linux kernel versions, resulting in a local denial of service. A prior regression removed a 'gratuitous' stack push in the non-kjump path; however, the purgatory code shipped by kexec-tools still reads above its stack top expecting that return address, triggering a memory fault that aborts the kexec reboot or crashes the system. No public exploit exists and EPSS sits at the 6th percentile, placing this firmly in the reliability-regression category rather than an adversarial attack surface.
NULL pointer dereference in the Linux kernel IOMMU subsystem crashes affected systems during PCI device reset when a default domain allocation has previously failed. Local users with low-privilege access on Linux kernel versions in the 7.0 branch (prior to 7.0.10) and before 7.1 can trigger a kernel panic through the `pci_dev_reset_iommu_done()` function, resulting in a denial of service. No public exploit code exists and no active exploitation has been confirmed; EPSS of 0.15% (5th percentile) reflects the low real-world exploitation likelihood given the edge-case hardware precondition and local-only attack vector.
Indefinite kernel hang in the Linux kernel's drm/gma500 Oaktrail LVDS display initialization code can be triggered locally on systems with Intel GMA 500/Oaktrail graphics hardware, causing a denial of service requiring a hard reboot. The defect lies in error handling that incorrectly attempts to deregister an I2C adapter obtained via i2c_get_adapter() - which only increments a reference count - rather than restricting cleanup to adapters the driver itself allocated. No public exploit code exists and EPSS is 0.17% (7th percentile), consistent with a niche, largely retired hardware platform.
NULL pointer dereference in the Linux kernel arm_mpam subsystem allows a local low-privileged attacker to crash the kernel by triggering a specific cleanup code path before initialization completes. The flaw exists in __destroy_component_cfg(), which unconditionally dereferences the embedded 'garbage' structure to free the configuration array even when mpam_disable() is called before __allocate_component_cfg() has ever run. No public exploit code exists and the EPSS score of 0.15% (5th percentile) confirms negligible real-world exploitation activity; vendor-supplied patches are available via upstream stable commits.
Signature type-binding bypass in @sigstore/core (npm) allows an attacker who controls the `payloadType` field of a DSSE envelope to substitute every ASCII character with a Unicode variant whose low byte matches, producing PAE bytes identical to a legitimate signature and causing verification to pass for a mismatched content type. All versions up to and including 3.2.0 are affected; a working proof-of-concept is included in the GitHub security advisory GHSA-jfc7-64v2-mr8c. The vulnerability is not confirmed in the CISA KEV catalog, but exploit code is publicly available, making exploitation straightforward for any attacker positioned to craft or relay DSSE envelopes.
Use-after-free in Envoy's HTTP OAuth2 filter (envoy.filters.http.oauth2) across versions 1.37.0-1.37.4 and 1.38.0-1.38.2 enables unauthenticated remote attackers to crash Envoy worker threads by racing a connection teardown against an in-flight async token exchange. When a downstream client disconnects while the filter is awaiting an async OAuth2 token response, the late AsyncClient callback invokes StreamDecoderFilterCallbacks on an already-freed object, producing undefined behavior and worker crashes. The reporter explicitly disclaims RCE - confirmed impact is availability loss (DoS); allocator-dependent memory corruption effects beyond crash are theoretically possible but not demonstrated. No public exploit or CISA KEV listing identified at time of analysis.
Use-After-Free crash in Envoy's ext_authz HTTP filter allows remote unauthenticated attackers to trigger a segmentation fault and process crash by rapidly establishing and tearing down downstream connections when per-route authorization overrides are configured. Affected are Envoy versions 1.36.0-1.36.8, 1.37.0-1.37.4, and 1.38.0-1.38.2. No public exploit code has been identified and this vulnerability is not listed in the CISA KEV catalog, though the race condition exploitability is constrained by requiring specific ext_authz per-route override configuration.
PROXY Protocol v2 TLV overflow in Envoy Proxy versions 1.34.0 through the pre-fix series allows an operator-level attacker on an adjacent network to smuggle extraneous bytes into upstream requests. Envoy's header generator writes TLV content exceeding the protocol-mandated 65535-byte maximum without adjusting the length field, creating a header where declared length and actual byte count diverge - a textbook CWE-130 length-parameter inconsistency. No public exploit code has been identified and this vulnerability does not appear in the CISA KEV catalog; risk is constrained by the high-privilege and adjacent-network prerequisites reflected in the CVSS 4.8 medium score.
Envoy proxy's External Processing (ext_proc) filter crashes with a use-after-free when an ext_proc gRPC server sends a single batched gRPC message containing multiple specially crafted ProcessingResponse objects. Affected versions span 1.34.0 through pre-fix releases across four maintained branches; fixed releases are 1.35.13, 1.36.9, 1.37.5, and 1.38.3. With CVSS PR:L, exploitation requires control of or compromise of the ext_proc gRPC server endpoint, limiting the realistic attacker population but making this a high-priority patch for any deployment running ext_proc. No public exploit code and no CISA KEV listing exist at time of analysis.
Certificate SAN validation in Envoy Proxy is bypassed when an attacker-controlled upstream serves a TLS certificate containing an embedded NUL byte in the dNSName Subject Alternative Name field. Envoy versions prior to 1.35.11, 1.36.7, 1.37.3, and 1.38.1 are affected, allowing an attacker who controls or can impersonate an upstream TLS endpoint to pass Envoy's configured SAN match check with a fraudulent certificate and intercept proxied traffic. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog.
The OAuth2 HTTP filter in Envoy Proxy prior to versions 1.35.11, 1.36.7, 1.37.3, and 1.38.1 implements AES-256-CBC encryption for the PKCE CodeVerifier cookie without any authentication tag, creating a classic padding oracle through differential HTTP responses on the /callback endpoint. An attacker who obtains the victim's encrypted CodeVerifier cookie and a stolen authorization code can recover the plaintext PKCE code_verifier in approximately 6,200 crafted requests (~100 seconds), then complete the OAuth token exchange to hijack the victim's access token. No public exploit code or CISA KEV listing has been identified at time of analysis; vendor-confirmed patches are available in all four current release branches.
Network annotation injection in KubeVirt's VirtualMachineInstance API allows authenticated tenants to attach launcher pods to arbitrary network namespaces by writing unsanitized JSON into the Multus CNI default-network annotation. Exploitation is gated on the ExternalNetResourceInjection Beta feature gate being explicitly enabled by a cluster-admin (off by default, first available in OpenShift Virtualization 4.21), which intentionally bypasses the NAD namespace lookup that would otherwise reject malformed annotation values. A successful attack permits cross-namespace network attachment and IP/MAC address impersonation on segments normally segregated from tenant workloads; no public exploit has been identified at time of analysis.
Cross-site scripting in jupyter/nbconvert versions 7.17.0 and earlier allows any user with notebook write access to inject and execute arbitrary JavaScript in the browsers of users who view HTML-exported notebooks. The flaw stems from the `data_mermaid` rendering block in `share/templates/lab/base.html.j2`, which outputs `text/vnd.mermaid` cell content directly into HTML without escaping, enabling tag breakout from the enclosing `<pre>` element. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in the CISA KEV catalog; however, the stored-XSS pattern means any user who views a maliciously crafted HTML export is at risk.
Use-after-free memory corruption in Google Chrome's Payments component on Android (prior to 149.0.7827.201) enables a local attacker with physical access to the device to trigger heap corruption, yielding high impact across confidentiality, integrity, and availability. The physical-access requirement (CVSS AV:P) substantially constrains the exploitable population to scenarios such as unattended or stolen devices. No public exploit identified at time of analysis, and no CISA KEV listing exists, indicating this has not been observed in active exploitation campaigns.
Lockdown mode in GitHub MCP Server HTTP deployments (versions 0.22.0-1.1.1) fails to isolate per-user GraphQL credentials due to a process-global singleton that retains the first authenticated user's client for the lifetime of the process. All subsequent users' IsSafeContent checks-the security gate that decides whether content from external contributors is sanitized before reaching the AI model-are evaluated under the first user's GitHub token and identity, producing wrong access control decisions for every user except the first. A publicly available proof-of-concept confirms the singleton pointer collision; no active exploitation (CISA KEV) has been confirmed at time of analysis.
Arbitrary filesystem write via zip-slip in K3s etcd snapshot restoration affects all K3s versions prior to 1.35.3+k3s1, 1.34.6+k3s1, and v1.33.10+k3s1. A maliciously crafted zip archive with path-traversal entries in member filenames (e.g., '../../etc/cron.d/evil') can overwrite arbitrary files on the host filesystem when an administrator performs a compressed etcd snapshot restore. No public exploit code has been identified at time of analysis, and this vulnerability is not listed in CISA KEV.
Decompression bomb (zip bomb) in Trivy's Helm chart scanner causes denial of service via OOM kill when processing a crafted .tgz archive. Affected versions prior to 0.71.0 use io.ReadAll without any size cap on tar entries inside Helm chart archives, allowing a small compressed file to expand to gigabytes in memory. An attacker who can place a malicious archive in any path that Trivy is directed to scan can reliably crash the Trivy process. No public exploit or CISA KEV listing at time of analysis.
Stack out-of-bounds write in Vim's spell suggestion engine allows a crafted `.spl`/`.sug` file pair to corrupt the call stack and crash the editor when spell suggestions are invoked. All Vim releases prior to 9.2.0653 are affected, with the flaw in `tree_count_words()` and `sug_filltree()` inside `src/spellfile.c`, where three fixed-size MAXWLEN-element stack arrays are indexed by an unbounded depth counter. No public exploit identified at time of analysis (CVSS 4.0 E:U), though the upstream fix commit includes a detailed proof-of-concept test case reproducing the out-of-bounds write.
Out-of-bounds read in Vim's text property subsystem allows a crafted undo file to crash the editor and potentially disclose adjacent heap memory. All Vim releases prior to 9.2.0670 are affected; the vulnerability is reachable whenever a user loads a malicious undo file with the persistent-undo feature compiled in and enabled. No public exploit or CISA KEV listing exists at time of analysis, placing real-world risk in the low-to-medium range despite the local crash severity.
Vim crashes with an out-of-bounds read when opening a maliciously crafted file encrypted with the xchacha20poly1305 cipher (VimCrypt~04! or VimCrypt~05!) whose body is shorter than a libsodium secretstream header. An unsigned integer underflow in the length calculation inside crypt_sodium_buffer_decode() causes the subsequent crypto_secretstream_xchacha20poly1305_pull() call to read far past the end of the input buffer, resulting in a denial-of-service crash. No public exploit code and no CISA KEV listing have been identified at time of analysis; however, the fix commit includes a concrete reproducer test case confirming the crash path.
Out-of-bounds heap read in Vim 9.2.0320-9.2.0678 allows a crafted undo or swap file to disclose adjacent heap memory or crash the editor when a victim opens it. The flaw resides in the text property (textprop) subsystem: a maliciously encoded virtual-text property with an oversized tp_text_offset is converted directly to a heap pointer without bounds validation in both the undo-restoration path (um_goto_line) and the display path (get_text_props). The CVSS 4.0 score of 6.8 (Medium) reflects local-only exploitation requiring active user interaction; no public exploit is confirmed at time of analysis, though the fix commit includes detailed test scaffolding constituting a near-complete attack blueprint.
Stack out-of-bounds write in Vim's spell_soundfold_sofo() function (src/spell.c) allows local exploitation when a SOFO-based spell language is active and the soundfold path is triggered with a word exceeding MAXWLEN (254) bytes, corrupting the call frame and crashing the editor with theoretical code execution potential. All Vim releases prior to 9.2.0698 carrying the single-byte SOFO branch are affected; the vendor has released a confirmed fix at v9.2.0698. No public exploit code exists and no CISA KEV listing is present; the CVSS 4.0 E:U supplemental metric confirms exploitation remains unproven at time of analysis.
Out-of-bounds read in Nokogiri's `NodeSet#[]` method (all versions before 1.19.4) enables an attacker who can supply a large negative integer index to bypass a 32-bit-truncated bounds check and trigger a read outside the node set's allocated memory. On CRuby (MRI Ruby), this typically crashes the process - denial of service - with ancillary potential for heap memory disclosure; on JRuby, the runtime's managed memory prevents memory corruption, but an incorrect node is silently returned. No public exploit or active exploitation has been identified; exploitation requires the target application to forward attacker-controlled integers directly to the `NodeSet#[]` or `#slice` API.
DNS response manipulation in PowerDNS Recursor 5.4.x allows network-positioned attackers to inject false DNS records through insufficient validation of answers received from authoritative servers, resulting in low-integrity violations in name resolution for downstream clients. The 5.4.x branch received hardening patches per PowerDNS advisory 2026-08 to close this validation gap. No public exploit has been identified at time of analysis and the vulnerability is not listed in the CISA KEV catalog.
Denial of service in PowerDNS dnsdist's DNS over HTTP/3 (DoH3) handler allows unauthenticated remote attackers to exhaust server memory by opening high volumes of concurrent QUIC streams carrying crafted queries. Each crafted query triggers an exception that defers buffer deallocation until QUIC connection teardown rather than releasing memory promptly, enabling cumulative memory exhaustion across simultaneous streams. No public exploit code or active exploitation has been identified at time of analysis, and the attack surface is limited to instances with DoH3 explicitly enabled.
dnsdist's SetMacAddrAction handler exposes operators to uninitialized memory leakage in DNS responses and potential service crashes when the action is configured in the ruleset. The flaw is reachable over the network without authentication (AV:N/PR:N), but the high attack complexity (AC:H) constrains real-world impact to deployments that have explicitly enabled SetMacAddrAction - a non-default configuration. No public exploit code exists and no CISA KEV listing is present at time of analysis; the PowerDNS security team (Open-Xchange) reported this internally, suggesting responsible disclosure rather than observed active exploitation.
TCP connection resource leak in PowerDNS dnsdist allows remote unauthenticated attackers to exhaust backend TCP connection capacity by sending IXFR (Incremental Zone Transfer) queries. Affected dnsdist instances fail to promptly release outgoing TCP connections to backends after IXFR processing, leaving them open until OS-level timeout; under sustained query volume this can exhaust the backend's concurrent connection limit or the dnsdist process's file descriptor table. No public exploit has been identified at time of analysis, and no CISA KEV listing exists, but the AV:N/AC:L/PR:N attack surface makes the trigger trivially automatable from the internet.
DNSSEC validation in PowerDNS Recursor can be silently undermined by network-positioned attackers who spoof DNS replies to trick the Recursor into permanently marking a legitimate authoritative server's IP as EDNS-incapable. Once that state is poisoned, the Recursor is unable to perform proper DNSSEC validation for any zones served by that authoritative server, effectively disabling a critical security layer for downstream clients. No active exploitation has been confirmed by CISA KEV, and no public proof-of-concept has been identified at time of analysis; however, the attack complexity is rated High due to the spoofing prerequisite.
ZONEMD validation bypass in PowerDNS Recursor allows a cryptographically invalid DNS zone to be accepted as legitimate when ZoneToCache is configured with ZONEMD validation enabled. This undermines the integrity guarantee that ZONEMD is designed to provide - an attacker capable of serving a crafted zone to the resolver (e.g., via a rogue authoritative source or network interception) can cause the recursor to cache and serve tampered DNS data. No public exploit code has been identified and exploitation is not confirmed in CISA KEV; real-world risk is constrained by the non-default configuration prerequisite.