Nest is a framework for building scalable Node.js server-side applications. Prior to 11.2.4 and 12.0.2, a single message with a deeply nested object in its pattern can terminate a NestJS microservice using the TCP or RabbitMQ transport. ServerTCP#handleMessage and ServerRMQ#handleMessage pass a client-controlled non-string pattern to JSON.stringify to derive the handler lookup key; sufficiently deep nesting throws RangeError: Maximum call stack size exceeded, and the unhandled promise rejection terminates Node.js under its default behavior. An attacker who can reach the TCP port or publish to the consumed RabbitMQ queue or exchange can crash the service on demand; other transports are not affected because their patterns arrive as strings. This issue is fixed in versions 11.2.4 and 12.0.2.
CVE-2026-102006 is a vulnerability in Wind River VxWorks 7 prior to version 26.09 where certain system call arguments cause the process management subsystem to fail to release allocated kernel memory when terminating an application. This results in a memory leak affecting system availability. The issue is fixed in version 26.09.
Eleveo Call Recording Software version 9.7.0 contains an LDAP injection vulnerability in the User Management component, specifically in the /callrec/userAddAction.do file. This vulnerability allows remote attackers to manipulate the Username argument to perform LDAP injection. Public exploit code is available, but no vendor response or patch has been provided.
CVE-2026-101144 is a medium severity vulnerability in Eleveo Call Recording Software version 9.7.0. It involves improper access controls in the Query Builder component, specifically in the /callrec/searchAction.do file. The flaw allows remote exploitation without user interaction and requires low privileges. Public exploit code has been disclosed, but no vendor response or patch is currently available.
CVE-2026-102004 is a high-severity vulnerability in Wind River VxWorks 7 prior to version 26.09. It involves an out-of-bounds write caused by specific system call arguments leading to memory corruption in the memory management subsystem. This flaw can result in confidentiality, integrity, and availability impacts. The issue has been fixed in version 26.09.
CVE-2026-101141 is a medium severity cross-site scripting (XSS) vulnerability in Eleveo Call Recording Software version 9.7.0. The flaw exists in the /callrec/audio.jsp file within the Play Audio Page component, where manipulation of the viewRoleId or cfType arguments can lead to XSS. The vulnerability can be exploited remotely and requires low privileges with partial user interaction. The vendor has not responded to the disclosure, and no patch or remediation is currently confirmed.
The npm package 'native-runner' contains malicious code that, upon import, silently waits for and executes a hidden binary named node_runtime_helper.exe from a Windows system-like directory. This behavior occurs as a side effect of requiring the package, without any explicit API call. The package depends on 'img-to-native', which places the malicious binary on the system. This drop-and-execute chain can lead to full system compromise, and removal of the package does not guarantee eradication of the malware.
Cameron John Wagenius, a former US Army soldier, was sentenced to 70 months in prison for hacking into AT&T and Verizon systems between April 2023 and December 2024. He stole and publicly disclosed confidential call detail records of a government official and conspired with others to extort and defraud multiple organizations by using stolen credentials obtained via brute force attacks. The conspirators also sold stolen data on cybercrime forums and attempted to extort at least $1 million from victims. Wagenius pleaded guilty to wire fraud conspiracy and extortion charges and was ordered to pay restitution. His accomplices were involved in other significant hacking campaigns affecting hundreds of organizations.
WolfSSL versions 5.8.4 through 5.9.2 with certain build configurations have a vulnerability where a failed X509_verify_cert call can permanently add an unverified attacker CA to the shared CertManager. This flaw bypasses certificate validation for all consumers relying on the shared CertManager, including native TLS, OCSP, CRL, and direct certificate manager verification. The issue arises only when specific macros (OPENSSL_EXTRA, NO_CERTS, WOLFCRYPT_ONLY) are set in a particular way or when built with --enable-opensslextra, and the application explicitly calls X509_verify_cert.
Kyverno versions prior to 1.19.1 contain a server-side request forgery (SSRF) vulnerability. This issue arises because certain legacy API call paths do not enforce egress filtering or validate service URLs, allowing crafted policies or resource submitters to make Kyverno send HTTP requests to arbitrary internal or external hosts. The vulnerability can lead to unauthorized access to cloud metadata endpoints and internal services using Kyverno's network privileges. The flaw is fixed in version 1.19.1.
The Grav Data Manager plugin (getgrav/grav-plugin-datamanager) versions 1.0.1 through 1.4.4 render stored data entries in the item-detail view (admin/templates/partials/item.html.twig) without escaping, applying Twig's `raw` filter — in some cases after a striptags('<br>') call that PHP's strip_tags() bypasses by preserving allowed tags together with their attributes. An unauthenticated visitor who submits a front-end form whose submissions are saved to user/data can store an HTML payload that executes as JavaScript in the session and origin of an administrator who later opens that entry in the classic admin panel, running with that administrator's privileges and CSRF token. Execution occurs without further interaction for list values (such as checkbox or multi-select fields) and on hover for ordinary text fields. Sites using the Grav 2.0 Admin Next interface are not affected, because it renders the same data through a separate, correctly escaping code path. The issue is fixed in Data Manager 1.4.5.
vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the WhisperFeatureExtractor and speech_tokenizer PreTrainedTokenizerFast loads in vllm/model_executor/models/funaudiochat.py and the Qwen2VLConfig.from_pretrained call used by Tarsier2ProcessingInfo in vllm/model_executor/models/qwen2_vl.py. As a result, deployments pinned to a reviewed revision still resolve these behavior-affecting processor, tokenizer, and config artifacts from the repository's default revision, so a later change to the upstream default branch can alter audio preprocessing, speech tokenizer behavior, or Tarsier2 configuration without any change to the operator's configured pin. This is a supply-chain integrity and reproducibility failure for pinned deployments; it is residual to the earlier fix tracked as GHSA-3ww4-5jv9-j5gm / CVE-2026-47155 and does not constitute remote code execution or a trust_remote_code=False bypass. The issue is fixed in version 0.28.0.
CVE-2026-100544 is a high-severity vulnerability in the openclaw voice-call package before version 2026.8.1. The flaw allows inbound callers admitted by the deployment's inbound-call policy to bypass authorization checks because the caller's identity and non-owner status are not propagated. This can cause owner-only tool filtering to fail, exposing the agent's normal tool authority to remote callers. Exploitation can lead to unauthorized reading of data, file modification, command execution, or control of connected services depending on the agent's configuration. The issue is fixed in version 2026.8.1.
SCBE-AETHERMOORE is a geometric AI governance and evaluation framework. Starting in version 4.0.2 and prior to version 4.2.1, the AetherBrowser API server (`scripts/aetherbrowser/api_server.py`) exposes the `POST /api/ops/check-email` endpoint without any authentication. Any remote attacker can call this endpoint and trigger execution of the `email_reader.py` subprocess, which connects to configured ProtonMail or Gmail accounts via IMAP and returns email metadata (sender, subject, body snippet) in the JSON response. The server binds to `0.0.0.0:8100` by default with CORS set to `allow_origins=["*"]`, making it reachable from any network or browser origin. Version 4.2.1 patches the issue.
RabbitMQ is a messaging and streaming broker. From 4.0.0 until 4.3.3, 4.2.9, 4.1.14, and 4.0.23, AMQP 1.0 management GET /bindings exposes full binding topology to any authenticated AMQP user without resource/management permission checks. the AMQP 1.0 HTTP-over-AMQP management endpoint GET /bindings (the rabbitamqpmanagement handler) enumerates bindings between an arbitrary source exchange and destination queue/exchange in the caller's virtual host without performing any resource-level permission check. Unlike every sibling operation in the same module (which call checkresourceaccess / bindingchecks), the GET handler ignores the authenticated User and returns the binding list unchanged. As a result, any authenticated AMQP 1.0 client that can open a management link pair , including users with no management/monitoring/policymaker/administrator tag , can enumerate the complete binding topology (source exchanges, destination queues/exchanges, routing keys, and binding arguments) of the virtual host they can access. The equivalent HTTP management API (GET /api/bindings) Confidentiality impact: a non-management AMQP 1.0 user can enumerate the complete routing topology of any virtual host it can connect to , every (source exchange, destination queue/exchange, routing key, binding arguments) This issue is fixed in versions 4.3.3, 4.2.9, 4.1.14, and 4.0.23.
A memory management vulnerability in the Linux kernel's rtl8723bs staging driver was fixed. The issue involved incorrect freeing of a vmalloc-allocated buffer using kfree(), which can cause memory corruption. The fix replaces the improper kfree() call with the correct vfree() via rtw_hal_data_deinit(). The vulnerability affects specific linux-azure kernel versions and has a medium severity rating.
A vulnerability in the Linux kernel's x86 Control Flow Integrity (CFI) implementation related to FineIBT hash offset handling has been fixed. The issue was caused by an incorrect offset used in the cfi_get_func_hash() function, leading to a mismatch in the hash read from the FineIBT preamble. This caused the kernel to fail CFI checks and crash when making the first indirect call into a BPF struct_ops program, such as during TCP congestion control initialization. The fix aligns the offset definitions to ensure consistent hash reading.
A vulnerability in the Linux kernel's BPF subsystem was resolved that involved improper marking of NULL call arguments in the check_func_arg() function. This flaw allowed certain scalar registers to be incorrectly pruned during checkpoint creation, potentially leading to incorrect behavior in BPF programs. The issue also affected the enforcement of parameters in bpf_get_local_storage(), similarly causing pruning against arbitrary scalar values. The vulnerability has been fixed by marking these arguments precisely to prevent such pruning.
A vulnerability in the Linux kernel's net: dsa: bcm_sf2 component allows a user without CAP_NET_ADMIN privileges to trigger a buffer over-read and potential memory corruption by providing a smaller buffer size than the number of installed CFP rules during an ETHTOOL_GRXCLSRLALL ioctl call. This occurs because the bcm_sf2_cfp_rule_get_all() function walks the entire cfp.unique bitmap without checking the caller's buffer size, leading to out-of-bounds memory access.