This vulnerability involves the accel/amdxdna component related to guard management mailbox channel cleanup, which may lead to a NULL pointer issue. The advisory provides no detailed technical information or impact description. The affected product is Microsoft software version 4.0. No CVSS score or exploit information is available.
Guard Core is an API security core engine for the FastAPI-Guard, FlaskAPI Guard, and DjAPI Guard adapters. From 3.5.0 until 3.13.0, built-in penetration-detection patterns in guard_core/handlers/suspatterns_handler.py execute through a direct pattern.search(content) call instead of the timeout-protected create_safe_matcher() path used for custom patterns. The built-in template-injection detector contains an unbounded expression with quadratic behavior when scanning repeated opening braces. Because penetration detection and enhanced mode are enabled by a stock SecurityConfig, an unauthenticated request can block the shared asyncio event loop, affecting all application endpoints. The triggering input does not produce a detection match, so the request can complete without incrementing ban counters or being recorded as suspicious. This issue is fixed in version 3.13.0.
Prime CMS (@primecms/core) through 0.4.0-beta.0 contains an authentication bypass vulnerability that allows unauthenticated attackers to register accounts via the createUser mutation on /prime/graphql. Because the isAuthenticated() guard has its check commented out and createAbility() hard-codes every role to admin, attackers can log in with full administrative control over content, users, webhooks and settings.
IBM ContextForge MCP Gateway 1.0.8 through 1.0.11, and v1.1.0 is vulnerable to denial of service due to an incomplete denylist in the jq filter guard module (CWE-184). An authenticated remote attacker can supply a crafted `jsonpath_filter` value containing `halt`, `halt_error`, or unbounded-loop built-ins that bypass the safety check, causing gateway worker process termination or sustained resource exhaustion.
Injection Guard plugin for WordPress versions before 1.4.0 has a cross-site request forgery (CSRF) vulnerability. This flaw allows unauthenticated attackers to trick a logged-in administrator into triggering a password hash export. The vulnerability arises because the ig_settings admin page processes GET parameters without nonce verification. Exploitation results in writing WooCommerce customer email and password hashes to an unprotected file, which can be downloaded without authentication for offline cracking.
A flaw was found in abrt. The completeness check introduced to mitigate CVE-2026-54228 and CVE-2026-54229 only verifies that a dump directory contains a "count" element, which an unprivileged user can create. This allows a local attacker to bypass the guard and again poison elements in, or seize ownership of, dump directories while privileged ABRT handlers are processing them. Related Problems2 mutating methods remain ungated by the same check.
Dockhand before 1.0.55 contains a path traversal vulnerability that allows authenticated users with stacks:edit to bypass the file-browser path guard by combining symlinks with '..' sequences. Attackers can read /proc/self/environ, the SQLite database, and the encryption key via /api/system/files/content, or move them using the stack relocate endpoint, exposing stored credentials.
Vvveb through 1.0.8.6 contains an SQL injection vulnerability because the @KEYS(:each) macro emits PHP array keys as backtick-quoted INSERT column names without escaping, and several procedures in admin/sql/mysqli/product.sql and admin_auth_token.sql omit the @FILTER guard. Authenticated back-office users able to save products can submit product_<feature> POST array keys containing backticks to redirect bound values into unexposed columns or break SQL statements.
MCUboot's Zephyr serial-recovery upload handler bs_upload() in boot/boot_serial/src/boot_serial.c decodes the host-supplied image field of an MCUmgr image-upload request into img_num (a full 32-bit value) and latches it on the off == 0 packet without any range check. It then passes img_num straight to flash_area_id_from_multi_image_slot(img_num, 0), which on the Zephyr port resolves to __flash_area_ids_for_slot() in boot/zephyr/include/sysflash/sysflash.h and indexes the static const int all_slots[] table as all_slots[img_num 2]. That table holds only MCUBOOT_IMAGE_NUMBER 2 entries, so any img_num >= MCUBOOT_IMAGE_NUMBER is an out-of-bounds read.
The path is reachable by anyone with access to the serial / SMP recovery console, which is unauthenticated by design, and the bad value is supplied before any image signature is verified. The defect is specific to the default non-direct upload path (MCUBOOT_SERIAL_DIRECT_IMAGE_UPLOAD disabled) on the Zephyr port with the array-backed slot layout (not CONFIG_SINGLE_APPLICATION_SLOT). It is not gated on a multi-image configuration: MCUBOOT_IMAGE_NUMBER defaults to 1, so even a default single-image build is affected for any image >= 1. The MCUBOOT_SERIAL_DIRECT_IMAGE_UPLOAD path is safe (its flash_area_id_from_direct_image() switch returns -EINVAL for unknown indices).
flash_area_id_from_multi_image_slot() takes a signed int image_index while img_num is uint32_t, so the attacker controls the sign of the index as well as its magnitude: an image of 0xFFFFFFFE converts to -2 and reads all_slots[-4], off the front of the table, while large positive values read off the end (CWE-192). (0xFFFFFFFF itself is the "no image given" sentinel and selects image 0.) Either way the index multiply img * 2 is performed in int, so a wild offset yields a bus/hard fault, giving a reliable denial of service of the bootloader during recovery. For moderately out-of-range indices the read returns adjacent .rodata as a garbage flash-area id that is handed to flash_area_open(); this normally fails with MGMT_ERR_EINVAL, but if the garbage id happens to match a valid non-image partition, the subsequent erase/write can target a partition outside the image set (CWE-787). The attacker-supplied value is never echoed back, so there is no information disclosure. The fix is to reject img_num >= BOOT_IMAGE_NUMBER on the off == 0 packet, before the slot lookup. The comparison must be made on the unsigned img_num: an equivalent guard applied after the conversion to int would not reject the negative case above. __flash_area_ids_for_slot() should bounds-check its own arguments as defence in depth.
This vulnerability was reported independently twice: by NVIDIA Project Vanessa on 2026-06-22, and by Kai Phan (@kaimandalic) on 2026-07-19 as GHSA-8v33-88p6-jffj, which is closed as a duplicate of this advisory. Both original reports are reproduced in full below.
A critical security vulnerability exists in Rundeck versions prior to 6.3.0-r0 and 6.1.0-r4 where applications evaluating Spring Expression Language (SpEL) expressions using SimpleEvaluationContext may be susceptible to a safety guard bypass when the SpEL expression compiler is active. This vulnerability can lead to high impact on integrity and availability without requiring user interaction or privileges.
bt_hci_le_per_adv_report_common() in subsys/bluetooth/host/scan.c handles the HCI_LE_Periodic_Advertising_Report (and _v2) event. It verified that the fixed event header was present and that the controller-declared Data_Length fits the tailroom of the periodic-sync reassembly buffer, but never checked that the event actually carried as many data bytes as evt->length claims. Both reassembly paths — DATA_STATUS_PARTIAL, and DATA_STATUS_COMPLETE while a chain is already in progress — then called net_buf_simple_add_mem(&per_adv_sync->reassembly, buf->data, evt->length), copying up to 255 bytes starting past the end of the event's real payload.
The Data_Length field originates in the HCI controller, so the flaw is reachable in host-only/split builds where the controller is a separate part reached over H4 UART, SPI, USB or an inter-core IPC transport: a malicious, compromised or buggy controller (or an attacker able to inject on that link) can send a report declaring a length larger than the event payload. The generic event dispatcher in subsys/bluetooth/host/hci_core.c only enforces a per-event minimum length and never cross-checks the subevent's internal length field, so nothing else catches it. Builds using Zephyr's own in-image controller are not affected, because the controller always emits a length matching the bytes it appends.
Exploitation additionally requires CONFIG_BT_PER_ADV_SYNC with CONFIG_BT_PER_ADV_SYNC_BUF_SIZE > 0 (the option defaults to 0, which compiles the reassembly block out; LE Audio broadcast-sink and Public Broadcast Profile applications that reassemble fragmented periodic advertising data set it), and an established periodic advertising sync.
The consequence is an out-of-bounds read of adjacent HCI RX-pool memory — including stale contents of previously received HCI events — which is copied into the reassembly buffer and then delivered to every registered bt_le_per_adv_sync_cb.recv listener as if it were received advertising data, so the disclosed memory becomes application-visible and may be forwarded off-device. The write side remains in bounds because the tailroom check uses the same length value, so there is no memory corruption; impact is limited to information disclosure and undefined behaviour on the over-read. The fix drops any report whose declared length exceeds the remaining event payload, matching the guard already present in the extended and legacy advertising report handlers.
CVE-2026-73619 is a moderate severity vulnerability in GitPython used by Red Hat products such as Red Hat Satellite and Red Hat OpenStack Platform. The flaw allows an attacker to read arbitrary files on the filesystem by bypassing an incomplete denylist in the unsafe_git_archive_options guard when invoking the Repo.archive() function. This leads to unauthorized information disclosure at the privileges of the process running GitPython. No fix has yet been released for affected Red Hat products.
Jan versions from 0.8.5 up to but not including 0.8.8 contain a vulnerability in the agent-tools gate.rs component that allows attackers to bypass workspace guards. This flaw enables writing files outside the intended workspace via dangling symlinks, potentially allowing attackers to create or modify sensitive files such as ~/.ssh/authorized_keys on the host system by supplying a malicious repository and using prompt injection.
The Photo Engine (Media Organizer & Lightroom) WordPress plugin versions up to and including 6.5.7 contain a stored cross-site scripting (XSS) vulnerability via the 'hierarchy[].name' parameter. Authenticated users with subscriber-level access or higher can inject malicious scripts that execute when other users view the affected pages. The vulnerability arises from insufficient input sanitization and output escaping, combined with an authentication mechanism that relies solely on a user meta token without proper capability checks. This token is exposed on admin pages due to a commented-out guard, increasing the risk of exploitation.
CrewAI has a Server-Side Request Forgery (SSRF) vulnerability due to flaws in its SSRF guard and HTTP client. Both components fail to properly detect malicious content in URLs because they check opposite content, allowing attackers to inject malicious URLs. This vulnerability can be exploited by tools such as the ScrapWebsiteTool to perform SSRF attacks.