Just published.

langchain4j-aideepin through 3.32.0 contains an insecure direct object reference vulnerability in CharacterMessageController that allows authenticated users to access other users' message references due to missing owner checks. Attackers can supply another user's message uuid to the knowledge-embedding-ref, memory-embedding-ref, or graph-ref endpoints to retrieve knowledge-base chunks, long-term memory text, and knowledge-graph data.

Auth & access

MrDoc through 1.1.1 contains a missing authorization vulnerability in the project_doc_sort and manage_project_doc_sort handlers that allows registered users to modify other users' documents by supplying arbitrary document IDs in sort_data. Attackers owning any project can reparent and reorder documents by sequential ID and set their status to draft or removed, hiding published documents.

Auth & access

VanBlog through 0.54.0 contains an information disclosure vulnerability that allows unauthenticated attackers to read private and hidden articles via the POST /api/public/article/:id endpoint. Attackers can submit any non-empty password for private articles lacking an effective password, or request hidden articles, since getByIdWithPassword never checks the hidden flag, exposing full article content.

Auth & access

Pig before 4.1.2 contains a missing authorization vulnerability in SysClientController that allows authenticated users without client-management permission to retrieve OAuth2 client secrets via GET /client/{clientId} and GET /client/export. Attackers can fetch clientSecret values for seed clients like pig, app, and daemon or export the client table, then obtain tokens using those clients' configured grant types.

Auth & access

PowerDNS-Admin before 2026.11.0 contains a missing authorization vulnerability in the apply_records handler for /admin/template/<template>/apply that lets any authenticated User-role account modify shared zone templates. Attackers can replace a template's record set by name, injecting attacker-chosen NS, MX, A or CNAME records into every zone later created from it.

Auth & access

Movim through 0.35.5 contains an authorization bypass vulnerability in Post::ajaxGetPost that allows authenticated users to read restricted posts by supplying server, node and nodeid values. Attackers who know item identifiers can retrieve cached community or contact-only posts from the pod-wide cache before the XMPP authorization check runs.

Auth & access Server

Bitpoll through 0.11.0 contains an insecure direct object reference vulnerability that allows unauthenticated attackers to read votes and comments from other polls via unscoped object lookups in views.py. Attackers can pair any public poll slug with sequential vote or comment IDs in vote, vote_delete, vote_assign, or delete_comment GET requests to read names and comments from restricted polls.

Auth & access

Kubero 3.0.0 through 3.1.1 contains an authentication bypass vulnerability that allows unauthenticated remote attackers to forge administrative JWTs by using a hardcoded fallback signing secret when JWT_SECRET is unset. Attackers can sign tokens with arbitrary role, userGroups and permissions claims, which JwtStrategy.validate() trusts without a database lookup, to take control of pipelines, applications, users and API tokens.

Auth & access Server

Tinyproxy through 1.11.3 contains an access control bypass vulnerability in acl_string_processing() in src/acl.c that compares only trailing hostname characters without checking DNS label boundaries. Attackers controlling the PTR record for their source IP, such as untrusted.net, can satisfy an 'Allow trusted.net' entry to use the proxy despite deny-by-default ACLs.

Other

REBUILD before 4.5.3 contains an authorization bypass vulnerability in FeedsListController fetchDetails and fetchComments that skips privilege filtering and FeedsHelper.checkReadable checks. Authenticated attackers can enumerate sequential IDs via the feeds-details and comments-list endpoints to read private or team-scoped feed content, images, attachments and comments.

Auth & access

IETF Datatracker before 12.83.0 contains an authorization bypass vulnerability that allows NomCom role holders to access other committees' records by supplying sequential primary keys. Attackers can abuse view_feedback_nominee, view_feedback_topic, view_feedback_unrelated and edit_nominee to view confidential nominee identities, reclassify feedback, and change nominee email addresses.

Auth & access

Krayin CRM before 2.2.8 contains a broken object level authorization vulnerability in Lead TagController attach and detach actions that skip the data-scope check. Authenticated users restricted to individual or group data scope can add or remove tags on leads owned by users outside their scope.

Auth & access Web app

MikroDash before 0.8.79 contains a server-side request forgery vulnerability that allows any signed-in user, including the viewer role, to target internal addresses through notification channels. Attackers can set webhook or mail destinations to loopback, LAN or link-local addresses and call POST /api/notify-channels/{id}/test to probe the internal network.

Disclosure Server

frab through 0.9.1 contains a privilege escalation vulnerability that allows authenticated crew users to grant themselves orga roles by submitting unchecked nested conference_users_attributes. Attackers can abuse UsersController#update or #create, which skip filter_conference_users, to gain orga on any conference and fully manage its events, schedule, reviewers and crew.

Auth & access

DollarDollar through 4.4 contains a missing authorization vulnerability in add_expense() that allows authenticated users to modify other users' account balances by supplying foreign account_id values. Attackers can submit sequential account_id, destination_account_id, and group_id values to /add_expense to tamper with balances and inject expenses into foreign groups with an arbitrary payer.

Auth & access

Omnom through 0.9.0 contains a server-side request forgery vulnerability that allows authenticated users to make server requests to arbitrary hosts by submitting feed URLs to /add_feed. Attackers can supply internal addresses that getFeedInfo, gofeed ParseURL, and fetchImageAsInlineURL fetch without validation, using reflected error messages to probe internal hosts and ports.

Disclosure Server

Pico through 4.1.2 contains a server-side request forgery vulnerability in the feeds service that allows signed-up users to make the server fetch arbitrary URLs via the run SSH command. Attackers can upload a feeds digest post listing internal URLs and trigger immediate fetching through Fetcher.ParseURL to reach loopback, RFC1918, and cloud metadata endpoints.

Disclosure Server

OmniAgent through commit 12b98c5 contains an approval bypass vulnerability in ReflexionAgent._is_safe_bash_command that treats any command starting with find as safe without inspecting -execdir arguments. Prompt-injected agents can run find with -execdir to invoke binaries like dd, nc, perl or openssl through an unsandboxed shell, overwriting files or executing code without human approval.

Other

Munin munin-httpd 2.99.1 through 2.999.18 contains a path traversal vulnerability in the static file handler of Munin::Master::HTML::handle_request that allows unauthenticated attackers to read files outside the static directory. Attackers can request /static/../<file> to bypass the '/../' substring check and read munin.conf and plugin-conf.d files containing monitored service credentials.

Disclosure Web app

Universal-modder before commit e88f6876745427321eb755ebc98276063a962d8a contains a missing WebSocket Origin validation vulnerability in the HostLink class of its passthrough example mods that allows remote attackers to execute Minecraft commands. Any web page open in the player's browser can connect cross-site to the unauthenticated loopback WebSocket and send cmd messages that WorldBridge.command() runs with operator authority.

Other Browser & mail

Mediary Scout before 2026.11.05 contains a server-side request forgery vulnerability that allows any authenticated user, including non-owner accounts, to set an arbitrary PanSou base URL via savePanSouBaseUrlAction without role checks. Attackers can trigger probePanSou() to send server-side POST requests to internal hosts and use distinct error responses as an oracle for internal host and port discovery.

Disclosure Server

Storyden through 1.26.16 contains a privilege escalation vulnerability that allows users holding MANAGE_ROLES to grant the ADMINISTRATOR permission without being administrators. Attackers can call AccountAddRole on another controlled account or use RoleCreate and RoleUpdate to create administrator roles, taking full control of the instance.

Auth & access

Tencent BrowserSkill through 0.4.2 contains a missing authentication vulnerability that allows local attackers to impersonate the bsk daemon because remoteSocket() connects to the loopback WebSocket without credentials or peer verification. Attackers who bind the daemon port before it starts or after it exits receive the extension's full tool surface, including evaluate, to run JavaScript through CDP in logged-in browser tabs.

Other Browser & mail

Seahub through 14.0.10 contains a weak randomness vulnerability in MultiShareLinksBatch.post() that generates batch share link passwords using Python's non-cryptographic random.choices(). Authenticated users permitted to create share links can request many generated passwords via the multi-share-links batch endpoint, recover the Mersenne Twister state, and predict other users' share link passwords.

Other

A vulnerability was found in SourceCodester Smart Attendance System with QR Code Scanner 1.0. Affected is an unknown function of the file api/attendance_handler.php of the component JSON Endpoint. Performing a manipulation of the argument student_code results in missing authentication. It is possible to initiate the attack remotely. The exploit has been made public and could be used.

Auth & access Web app

A vulnerability was determined in SourceCodester Smart Attendance System with QR Code Scanner 1.0. Affected by this vulnerability is an unknown functionality. Executing a manipulation can lead to cross-site request forgery. It is possible to launch the attack remotely. The exploit has been publicly disclosed and may be utilized.

Auth & access

A security flaw has been discovered in SourceCodester Smart Attendance System with QR Code Scanner 1.0. This affects an unknown part of the file login.php. The manipulation results in session fixiation. The attack can be launched remotely. The exploit has been released to the public and may be used for attacks.

Other Web app

KUSANAGI provided by GMO Prime Strategy Co., Ltd. contains an HTTP header injection vulnerability that may allow arbitrary HTTP headers to be injected into the HTTP response when a user accesses a specially crafted URL.

Other

A vulnerability has been found in SourceCodester Smart Attendance System with QR Code Scanner 1.0. This impacts the function getMessage of the file create_admin.php of the component Admin Account Creation. Such manipulation of the argument e leads to missing authentication. The attack may be performed from remote. The exploit has been disclosed to the public and may be used.

Auth & access Web app SourceCodester Smart Attendance System With QR Code Scanner

An unauthenticated SQL injection in the element search endpoint allows stacked query execution, enabling escalation to an administrator.

Injection Web app Craftcms Cms

A flaw has been found in SourceCodester Smart Attendance System with QR Code Scanner 1.0. This affects an unknown function of the file reset_password.php of the component Password Reset. This manipulation of the argument username/password causes weak password recovery. The attack is possible to be carried out remotely. The exploit has been published and may be used.

Other Web app SourceCodester Smart Attendance System With QR Code Scanner

Mongoose is a MongoDB object modeling tool designed to work in an asynchronous environment. Prior to 6.13.11, 7.8.12, 8.24.3, and 9.9.2, Mongoose countDocuments() does not apply the configured sanitizeFilter query selector injection defense. An application that enables sanitizeFilter and passes attacker-controlled filter values to countDocuments() can therefore process query operator objects as operators rather than literal values, causing the count to include documents that should not match and potentially undermining authorization or application logic based on that count. This issue is fixed in versions 6.13.11, 7.8.12, 8.24.3, and 9.9.2.

Injection Automattic Mongoose

Mongoose is a MongoDB object modeling tool designed to work in an asynchronous environment. From 6.0.0 until 6.13.12, 7.8.13, 8.24.7, and 9.11.4, Mongoose stores the connection-level sanitizeFilter option in Connection.config, but Query.prototype._castConditions() and query casting check other option locations and do not read that configuration. An application that enables sanitizeFilter through mongoose.connect() or createConnection() and passes an attacker-controlled object into a query filter therefore receives a connection that reports the defense as enabled while query selector objects remain unsanitized. A crafted password value using a query operator can match an account record without the expected literal password, while global configuration through mongoose.set('sanitizeFilter') and per-query Model.find().sanitizeFilter() remain unaffected. This issue is fixed in versions 6.13.12, 7.8.13, 8.24.7, and 9.11.4.

Injection Automattic Mongoose

Mongoose is a MongoDB object modeling tool designed to work in an asynchronous environment. Prior to 6.13.12, 7.8.13, 8.24.7, and 9.11.4, Schema.prototype.virtual() builds nested virtual paths without applying the special-property validation already used by Schema.path() and Schema.add(). An attacker who can influence virtual field names in dynamic schemas, content-management form builders, or schema-as-code systems can supply path segments such as __proto__, constructor, or prototype, causing the virtual-path reducer to write enumerable properties to Object.prototype or Function.prototype. The resulting prototype pollution can spoof inherited property checks such as authorization flags and can provide function-prototype gadget state to other server-side JavaScript code. This issue is fixed in versions 6.13.12, 7.8.13, 8.24.7, and 9.11.4.

Other Server Automattic Mongoose

Unsafe reflection in the error mitigation schema validation in Amazon Braket Schemas before 1.32.2 might allow a remote authenticated user with S3 write access to the quantum task output location to execute arbitrary code on any machine that retrieves the quantum task results via a crafted task result document. We recommend you upgrade to amazon-braket-schemas version 1.32.2 or later.

Code execution Amazon amazon-braket-schemas

Relative path traversal in the local hybrid job input data download in Amazon Braket SDK before 1.127.7 might allow a remote authenticated user with S3 write access to a hybrid job's input data location to overwrite arbitrary files with the privileges of the user running the job via crafted S3 object keys. We recommend you upgrade to amazon-braket-sdk version 1.127.7 or later.

Disclosure Library AWS amazon-braket-sdk

Mongoose is a MongoDB object modeling tool designed to work in an asynchronous environment. Prior to 6.13.12, 7.8.13, 8.24.7, and 9.11.4, Mongoose can descend into inherited properties while loading database documents in init(), setting dotted paths in $__set, and merging update documents in utils.mergeClone. Attacker-controlled keys such as toString or valueOf can therefore cause Mongoose to write into shared built-in functions instead of an own property. A crafted document, Mixed path, non-strict document or subdocument, findOneAndReplace() replacement, chained update, or Query.prototype.merge() source can overwrite Object.prototype.toString.call, causing later queries on unrelated models to fail until the process restarts. Separately, when an attacker controls the dotted path passed to doc.set(path, value) for a subdocument configured with strict: false, descent through $__set can reach $__schema and alter shared validators or schema options for later documents. Mongoose already blocks __proto__, constructor, and prototype, so this issue overwrites properties on reachable shared objects rather than adding new properties directly to Object.prototype. This issue is fixed in versions 6.13.12, 7.8.13, 8.24.7, and 9.11.4.

Other Server Automattic Mongoose

Mongoose is a MongoDB object modeling tool designed to work in an asynchronous environment. Prior to 6.13.12, 7.8.13, 8.24.7, and 9.11.4, Mongoose uses attacker-controlled discriminator key values, including the default __t field, to index plain-object registries such as Model.discriminators and schema.discriminators without requiring an own property. Values such as constructor, toString, or hasOwnProperty can therefore resolve to inherited built-in functions instead of a registered discriminator, causing type confusion in model construction, document creation, hydration, query and update casting, embedded discriminators, and populate operations. In the bulkWrite() insertOne path with skipValidation enabled, a constructor value can cause a raw input object to be stored without schema casting or strict filtering, and a stored poisoned document can make every query that hydrates it fail until the document is removed. The weakness reads inherited properties but does not modify Object.prototype and is not prototype pollution. This issue is fixed in versions 6.13.12, 7.8.13, 8.24.7, and 9.11.4.

Memory safety Device Automattic Mongoose

Monoceros versions before 0.30 for Perl allow a request declaring a large Content-Length to kill a worker by exhausting memory. The request body is read in a loop that asks sysread for all of the remaining declared length at once. Perl grows the read buffer to that size before any data arrives, so a length larger than available memory makes the worker exit with an uncatchable out of memory error before the application is called. A remote attacker can force a worker restart with each request of a few bytes.

Other

A use-after-free issue was addressed with improved memory management. This issue is fixed in iOS 18.7.12 and iPadOS 18.7.12, iOS 26.7.3 and iPadOS 26.7.3, iOS 27.2.1 and iPadOS 27.2.1, macOS Golden Gate 27.2.1, macOS Sequoia 15.8.3, macOS Tahoe 26.7.3, visionOS 27.2.1, watchOS 26.6.1, watchOS 27.2.1. An attacker in Bluetooth range may be able to corrupt process memory.

Memory safety OS & kernel

Novu provides an API for sending notifications through multiple channels. Prior to 3.17.0, validateUrlSsrf in libs/application-generic/src/utils/ssrf-url-validation.ts uses the isPrivateIp deny-list without covering the RFC6598 100.64.0.0/10 shared address space or its IPv4-mapped IPv6 forms. The workflow HTTP request step and webhook filter condition can therefore send server-side requests when a configured literal address or resolved hostname falls in that range. In Alibaba Cloud, 100.100.100.200 may expose instance metadata, while self-hosted and multi-tenant deployments may expose provider-internal services, private infrastructure, service meshes, or VPN resources reachable through CGNAT addressing. This issue is fixed in version 3.17.0.

Disclosure Network gear Novuhq Novu

Novu provides an API for sending notifications through multiple channels. Prior to 3.15.0, the conditions filter webhook in libs/application-generic/src/usecases/conditions-filter/conditions-filter.usecase.ts passes the user-configured child.webhookUrl directly to axios.post without calling validateUrlSsrf. Evaluating a workflow condition can therefore make the worker send a POST request to a private or link-local service and return the response data for condition evaluation. The response is stored in execution details and exposed through GET /execution-details, which can disclose credentials from POST-capable metadata services or data from internal services, although metadata endpoints that require GET are not reachable through this path. This issue is fixed in version 3.15.0.

Disclosure Novuhq Novu

The XRootD project provides a high-performance, fault-tolerant, and secure solution for handling large amounts of data distributed across multiple storage resources. Prior to version 5.9.2, XRootD's rpCheck() function in src/XrdXrootd/XrdXrootdXeq.cc rejects a parent-directory component only when two dots are followed by a slash and does not reject a terminal two-dot component. An unauthenticated attacker can append /.. to an exported path in xrdfs ls or an HTTP PROPFIND request, bypass the all.export directory scope, and cause directory and filename enumeration in the parent directory under oss.localroot. The demonstrated vector does not permit file download or disclose file contents. This issue is fixed in version 5.9.2.

Other Xrootd

djust provides Phoenix LiveView-style reactive server-side rendering for Django with Rust-powered performance. Prior to versions 1.0.10 and 1.1.2, djust keys context output-safety grants by variable name and fails to revoke a grant when a template bind replaces the value associated with that name. A template using {% with %}, {% for %}, {% include ... with %}, tuple unpacking, or assign tags can therefore rebind a name that previously held mark_safe content and emit the new attacker-controlled value without escaping, including descendant lookups through the rebound name. This can cause stored or reflected cross-site scripting when a victim renders the affected template. This issue is fixed in versions 1.0.10 and 1.1.2.

XSS Web app djust-org Djust

The NXP ENET-QOS Ethernet driver's receive work handler eth_nxp_enet_qos_rx() in drivers/ethernet/eth_nxp_enet_qos hands each processed descriptor back to the DMA by writing desc->read.control = rx_desc_refresh_flags, which sets OWN_FLAG and BUF1_ADDR_VALID_FLAG. RDES0 of a receive descriptor is the DMA buffer address in read format, but the hardware write-back aliases that word with the VLAN tag fields (or, for a context descriptor, the timestamp low word), so it is destroyed on every completion. The success path rewrites desc->read.buf1_addr before re-arming; the four drop paths — errored or non-first descriptor, net_pkt_rx_alloc() failure, invalid descriptor length, and net_pkt_get_frag() failure — did not, re-arming the ring slot with the write-back residue as its DMA target address. The drop paths are driven entirely by received traffic and need no privilege or special configuration. With the defaults for this driver (CONFIG_NET_BUF_DATA_SIZE of 128 and 16 receive descriptors) a single full-size frame consumes twelve ring slots and twelve K_NO_WAIT fragment allocations, so a short burst of frames sent to the device exhausts the receive pools and reaches the allocation-failure paths; a frame with a receive error, or any frame whose first descriptor was already dropped, reaches the non-first-descriptor path. Once a slot has been poisoned, the DMA writes the next frame received in that slot to an address derived from the frame's VLAN tag fields rather than to a receive buffer. On the supported MCX SoCs that address falls in flash or reserved space, so the transfer raises a fatal bus error that the driver only logs — there is no reset or recovery path — and reception stops permanently until the device is rebooted; where the write is silently discarded instead, the untouched buffer is later handed to the network stack with a length, delivering stale buffer contents as if they were a received frame. A single triggering event is enough to make the condition persistent, so an attacker able to send traffic to the device can permanently disable its network interface.

Memory safety OS & kernel Zephyrproject Zephyr

The NXP ENET QoS Ethernet driver in drivers/ethernet/eth_nxp_enet_qos serialises transmission with a binary semaphore, data->tx.tx_sem, taken in eth_nxp_enet_qos_tx() and returned only by the transmit-completion interrupt handler tx_dma_done(). A frame that has been handed to the transmit DMA when the Ethernet link goes down is never written back, so that completion interrupt never arrives. The PHY callback eth_nxp_enet_qos_phy_cb() reacts to carrier loss only by calling net_eth_carrier_set(); it does not stop the DMA, take the descriptors back or release the outstanding frame, and no timeout exists to do so. The trigger is a link-down/link-up transition that lands while a frame is under the transmit DMA. It can be produced by anyone who controls the link partner or the cable — for example by administratively downing and restoring the attached switch port, or by restarting auto-negotiation from a peer device — and can be made reliable by generating traffic that keeps the device transmitting and repeating the transition. No credentials or user interaction are required, and the trigger is at the link layer rather than over IP. Because the semaphore is never given back, every subsequent call to eth_nxp_enet_qos_tx() fails with -EBUSY for the remaining life of the boot; restoring the link does not recover the interface, since the link-up path only rewrites MAC_CONFIGURATION. The affected interface continues to receive but can no longer transmit anything — no ARP replies, no TCP acknowledgements, no DHCP renewals — until the device is reset. The net_pkt and the net_buf fragments of the abandoned frame are likewise never unreferenced and never return to their pools. The security-relevant property is persistence: a momentary link flap leaves the interface permanently unable to send, so the denial of service outlasts the attacker's access to the link. There is no confidentiality or integrity impact. The fix adds a claim/release pair around the outstanding packet so that completion and abandonment are mutually exclusive, aborts an outstanding frame from the PHY callback on carrier loss (stopping the DMA, flushing the MTL transmit queue and reinitialising the descriptor ring), and arms a transmit watchdog, CONFIG_ETH_NXP_ENET_QOS_TX_TIMEOUT_MS, so that any non-completing DMA — not only a carrier loss — reclaims the transmit slot and the packet.

Denial of service OS & kernel Zephyrproject Zephyr

When CONFIG_THREAD_STACK_MEM_MAPPED is enabled, each thread stack is mapped into the kernel virtual address space with a guard page on either side, and the mapping is released during abort cleanup. k_thread_abort_cleanup() in kernel/thread.c has two paths: a deferred one for self-abort, which stashes the mapping in thread_cleanup_stack_addr/thread_cleanup_stack_sz, and an immediate one for a thread aborted by some other thread. The immediate path called do_thread_cleanup(), which ignored its thread argument and only ever unmapped the stashed address — a value populated exclusively by the self-abort path. The mapped stack of a thread aborted by another thread was therefore never unmapped. Every k_thread_abort() of a non-current thread leaks the thread's stack region plus its two guard pages from the kernel virtual address space. Threads that end by returning from their entry function take the working self-abort path, so the leak is specific to explicit cancellation — k_thread_abort(), pthread_cancel(), osThreadTerminate(), or the kernel thread kill shell command. Both k_thread_create() and k_thread_abort() are system calls, so with CONFIG_USERSPACE an unprivileged user-mode thread that has been granted a thread object and a user-capable stack object can drive the leak from user mode, consuming a kernel-side resource it cannot otherwise touch. With CONFIG_KERNEL_VM_SIZE defaulting to 8 MB, a few hundred create/abort cycles exhaust the runtime mapping region. k_mem_map_phys_guard() then returns NULL, and map_thread_stack() guards that only with __ASSERT_NO_MSG(), which is compiled out of production builds; thread creation proceeds to write through a NULL-derived stack base and faults in kernel mode. The impact is a persistent denial of service — a kernel fatal error requiring a reboot — with no memory disclosure or corruption, and it is limited to configurations where memory-mapped thread stacks are enabled (x86 targets with X86_MMU and demand paging disabled, where the option is implied by default).

Denial of service OS & kernel Zephyrproject Zephyr

The HTTP client's debug helper print_header_field() in subsys/net/lib/http/http_client.c formatted an http_parser data-callback slice with a %s conversion: snprintk(output, len + 1, "%s", str). The parser hands its callbacks a pointer and a length into the caller's receive buffer, and those bytes are not NUL terminated. The size argument of snprintk() bounds only the destination, while the %s conversion measures the source first — lib/os/cbprintf_complete.c uses strlen() when no precision is given, and lib/os/cbprintf_nano.c uses strlen() unconditionally — so the read walks past the end of the header field and past the end of the receive buffer until it finds a zero byte. The slice reaches the helper from on_url(), on_header_field() and on_header_value(), which the parser invokes on bytes that http_wait_data() read straight off the socket into the application-supplied recv_buf. A malicious or compromised HTTP server, or an attacker able to answer a plaintext request, therefore controls both the header field and the bytes that follow it in the buffer, and can guarantee that no zero byte appears before the buffer ends, forcing the over-read. The in-tree tests/net/lib/http_client suite reproduces the over-read under CONFIG_ASAN=y on an ordinary GET. The impact is bounded. The whole helper body is compiled only when the HTTP client log module is explicitly set to Debug (CONFIG_NET_HTTP_LOG_LEVEL_DBG=y, which the global CONFIG_LOG_DEFAULT_LEVEL does not set), so default builds are unaffected. No out-of-bounds write occurs and none of the over-read bytes are emitted into the log, so there is no information disclosure; the consequence is an out-of-bounds read that can fault and panic the device where the buffer abuts an unmapped or MPU/MMU-protected region. The fix copies exactly the len bytes the parser offered with memcpy() and terminates them, instead of relying on a %s conversion (a %.*s precision would not have sufficed, since cbprintf_nano.c measures the source in full before applying it).

Memory safety Library Zephyrproject Zephyr

usbh_device_init() in subsys/usb/host/usbh_device.c rejects a device whose device descriptor advertises bNumConfigurations == 0, but it only logs the error and jumps to the cleanup label without setting the return code. At that point the local err is still 0 from the preceding successful descriptor read, so the function returns success while having skipped usbh_device_set_configuration() — the only place udev->cfg_desc is allocated. The device structure is zero-initialised by usbh_device_alloc(), so udev->cfg_desc remains NULL. usbh_device_connect() treats the zero return as a successful enumeration and calls usbh_class_probe_device(), which starts its descriptor walk from udev->cfg_desc. usbh_desc_get_next_function() passes that NULL pointer to usbh_desc_is_valid(), whose only guard compares the caller-supplied size, so head->bLength is read through a NULL pointer. The condition is triggered entirely by the attached peripheral: attaching a USB device (or any USB gadget emulating one) that reports bNumConfigurations = 0 reaches the defect directly from the UHC_EVT_DEV_CONNECTED_FS/_HS event handler in subsys/usb/host/usbh_core.c, with no authentication and no user interaction beyond the physical attachment. On targets where page zero is unmapped or CONFIG_NULL_POINTER_EXCEPTION_DETECTION is enabled this raises a fatal exception in the USB host thread, denying service to the host. Where address zero is readable, the descriptor walker instead parses low memory as a descriptor chain and can dispatch a class probe() with attacker-influenced values, causing further out-of-bounds reads through the still-NULL udev->cfg_desc in usbh_desc_get_iface(). No attacker-controlled write was identified, so the impact is availability plus a bounded out-of-bounds read. Note that CONFIG_USB_HOST_STACK is marked experimental.

Memory safety Zephyrproject Zephyr

The Xtensa MMU userspace port never initialized the per-thread page-table pointer thread->arch.ptables during thread setup, while arch_mem_domain_thread_add() in arch/xtensa/core/ptables.c uses a non-NULL value of that field as its "this thread is migrating from another memory domain" predicate. For a thread whose struct k_thread backing memory is not zeroed, the field holds stale data and the kernel takes the migration branch, passing that stale value to reset_region() as the base of a level-1 page table. The equivalent field on x86 is explicitly cleared in z_x86_userspace_prepare_thread() for exactly this reason; the Xtensa port adopted the predicate without the initialization. The condition is reachable from an unprivileged user-mode thread on a build with CONFIG_XTENSA_MMU, CONFIG_USERSPACE and dynamically allocated kernel objects. k_object_alloc(K_OBJ_THREAD) is a system call and allocates the thread object from the caller's resource pool without zeroing it, and the k_thread_create() verifier accepts such an object while forcing the K_USER option, so the migration branch is taken whenever the recycled heap bytes at that offset are non-zero. Because a user thread can also allocate, fill and free other kernel objects from the same heap, the stale bytes are groomable rather than merely random. Walking the bogus table in region_map_update() performs an unguarded index into l2_page_tables_counter[], may copy from and write through the attacker-influenceable level-1 base in dup_l2_table_if_needed(), and finally writes a page-table entry through a second pointer read out of that memory — once per page of the new thread's stack. The result is out-of-bounds reads and writes executed in supervisor mode, ranging from an immediate kernel panic (observed as Xtensa EXCAUSE 13 on Intel ADSP DSP firmware with user threads) to corruption of kernel memory and escape from the user-mode sandbox.

Memory safety OS & kernel Zephyrproject Zephyr