CVE Feed

HashiCorp Consul and Consul Enterprise are vulnerable to an authorization bypass in health check registration that may allow an authenticated user to modify health checks belonging to another service on the same node. This vulnerability (CVE-2026-115804) is fixed in Consul Community Edition 2.1.0 and Consul Enterprise 2.1.0 and 1.21.19.

OmniRoute 3.8.50 before 3.8.52 contains a missing authentication vulnerability that allows unauthenticated attackers to issue sync tokens via POST /api/sync/tokens when requireLogin is disabled. Attackers can use the issued token against GET /api/sync/bundle to retrieve decrypted provider credentials and plaintext API keys, with tokens persisting after login is re-enabled.

There is a reflected cross site scripting vulnerability in Esri Portal for ArcGIS versions 12.1 and prior which may allow a remote, unauthenticated attacker to create a crafted link which when clicked could potentially execute arbitrary JavaScript code in the victim’s browser. A patch is provided for ArcGIS Server software in General Availability and Extended support status, which includes versions 12.1, 12.0, 11.5, 11.3, and 11.1. Other versions are in mature or retired status, which are not patch eligible. Users working with unlisted versions are advised to upgrade and apply the patch.

There is a stored cross site scripting issue in Esri ArcGIS Server 11.1 and earlier on Windows and Linux that in some configurations allows a remote unauthenticated attacker to store files that contain malicious code that may execute in the context of a victim’s browser. A patch is provided for ArcGIS Server software in General Availability and Extended support status, which includes versions 12.1, 12.0, 11.5, 11.3, and 11.1. Other versions are in mature or retired status, which are not patch eligible. Users working with unlisted versions are advised to upgrade and apply the patch.

Esri ArcGIS Server versions 11.3 to 12.0 ONLY allows a remote, unauthenticated attacker to update service metadata. A patch is provided for ArcGIS Server software in General Availability and Extended support status, which includes versions 12.0, 11.5, and 11.3. Other versions are in mature or retired status, which are not patch eligible. Users working with unlisted versions are advised to upgrade and apply the patch.

There is s denial of service vulnerability in Esri ArcGIS Server versions 12.0 and prior that may allow a remote, unauthenticated attacker to supply crafted requests that may, under exceptional circumstances, consume all available worker threads and deny service to users. A patch is provided for ArcGIS Server software in General Availability and Extended support status, which includes versions 12.0, 11.5, 11.3, and 11.1. Other versions are in mature or retired status, which are not patch eligible. Users working with unlisted versions are advised to upgrade and apply the patch.

There is a reflected cross site scripting vulnerability in Esri Portal for ArcGIS versions 12.1 and prior which may allow a remote, unauthenticated attacker to create a crafted link which when clicked could potentially execute arbitrary JavaScript code in the victim’s browser. A patch is provided for impacted ArcGIS Server software in General Availability and Extended support status, which includes versions 11.5. Other versions are in mature or retired status, which are not patch eligible. Users working with unlisted versions are advised to upgrade and apply the patch.

There is a SQL injection vulnerability in Esri ArcGIS Server versions 12.1 and earlier on Windows and Linux that may allow a remote attacker with privileges to access the service (which could, depending on the service being configured in a non-default access configuration “public”, include an anonymous attacker) to issue database specific functions that should not be allowed. A patch is provided for ArcGIS Server software in General Availability and Extended support status, which includes versions 12.1, 12.0, 11.5, 11.3, and 11.1. Other versions are in mature or retired status, which are not patch eligible. Users working with unlisted versions are advised to upgrade and apply the patch.

HashiCorp Consul and Consul Enterprise are vulnerable to a denial of service in the HTTP API that may allow an unauthenticated attacker to exhaust memory on a Consul agent. This vulnerability (CVE-2026-105626) is fixed in Consul Community Edition 2.1.0 and Consul Enterprise 2.1.0 and 1.21.19.

A vulnerability was identified in FastStone Image Viewer up to 8.5. Affected is the function FUN_0062d96c of the component Box-Average Downsampler Engine. The manipulation leads to out-of-bounds read. It is possible to initiate the attack remotely. The vendor was contacted early about this disclosure but did not respond in any way.

Backstage is an open framework for building developer portals. Prior to 3.9.2 and from 4.0.0 until 4.0.2, DefaultCatalogRulesEnforcer in @backstage/plugin-catalog-backend compares the raw location.target URL against catalog.rules, while URL readers canonicalize the same URL before fetching it. An authenticated user with catalog.location.create permission can use percent-encoded path segments to make an untrusted URL match a trusted location pattern even though the canonical URL resolves outside that scope. This admits entity content from an untrusted source as trusted, while further downstream consequences depend on deployment configuration. This issue is fixed in versions 3.9.2 and 4.0.2.

HashiCorp Consul and Consul Enterprise are vulnerable to an authorization bypass in service mesh proxy registration. This may allow a user with service write permission to apply Envoy escape-hatch configuration without the required mesh write permission. This vulnerability (CVE-2026-105608) is fixed in Consul Community Edition 2.1.0 and Consul Enterprise 2.1.0 and 1.21.19.

When the vSolution Matrix feature is enabled, affected devices may make current screen content available through an unauthenticated RTSP/RTP stream to hosts on the same network segment. The stream is not encrypted, which may allow an unauthenticated host to view or capture presentation content where network access is not otherwise restricted. This vulnerability only affects devices with the vSolution Matrix feature enabled.

A vulnerability was determined in Omega Solution Restro POS up to 20261026. This impacts an unknown function of the file /sale-invoice/. Executing a manipulation of the argument productUnitSalePrice can lead to business logic errors. The attack may be performed from remote. The exploit has been publicly disclosed and may be utilized. The vendor was contacted early about this disclosure but did not respond in any way.

HashiCorp Consul and Consul Enterprise 1.17.0 up to 2.0.4 are vulnerable to an authorization bypass in the agent's service mesh leaf certificate API that may allow an attacker without a valid ACL token to retrieve a cached service leaf certificate and its private key. This vulnerability (CVE-2026-105765) is fixed in Consul Community Edition 2.1.0 and Consul Enterprise 2.1.0 and 1.21.19.

Open Neural Network Exchange (ONNX) is an open standard for machine learning interoperability. Prior to 1.22.0, _unpack_2bit() and _unpack_4bit() resize unpacked arrays according to attacker-controlled TensorProto.dims without verifying that raw_data or int32_data encodes the declared number of elements. Materializing an untrusted UINT2, INT2, UINT4, INT4, or FLOAT4E2M1 tensor through onnx.numpy_helper.to_array() or onnx.reference.ReferenceEvaluator can therefore expand a small packed payload into a disproportionately large allocation and exhaust memory. This issue is fixed in version 1.22.0.

Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in WP Chill Gallery PhotoBlocks photoblocks-grid-gallery allows Stored XSS.This issue affects Gallery PhotoBlocks: from n/a through 1.3.7.

Concurrent Execution using Shared Resource with Improper Synchronization ('Race Condition') vulnerability in VillaTheme Photo Reviews for WooCommerce woo-photo-reviews allows Leveraging Race Conditions.This issue affects Photo Reviews for WooCommerce: from n/a through 1.2.31.

Open Neural Network Exchange (ONNX) is an open standard for machine learning interoperability. From 1.4.0 until 1.22.0, ScanInferenceFunction, ScanInferenceFunction_opset8, and ScanInferenceFunction_opset9 do not safely validate the required num_scan_inputs attribute. A missing attribute is dereferenced as a null pointer, while a negative value is converted to size_t and can reach an exceptional oversized allocation path. An untrusted model can therefore terminate or disrupt shape inference, including a negative-value model that can pass onnx.checker.check_model(). This issue is fixed in version 1.22.0.

Missing Authorization vulnerability in VillaTheme Customer Coupons for WooCommerce woo-customer-coupons allows Exploiting Incorrectly Configured Access Control Security Levels.This issue affects Customer Coupons for WooCommerce: from n/a through 1.3.14.

Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in bPlugins Typing Text – Give your texts many typing effects text-typing allows Stored XSS.This issue affects Typing Text – Give your texts many typing effects: from n/a through 2.2.0.

Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in bPlugins Button Block button-block allows Stored XSS.This issue affects Button Block: from n/a through 1.2.7.

Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in bPlugins Content Slider Block content-slider-block allows Stored XSS.This issue affects Content Slider Block: from n/a through 3.2.1.

Open Neural Network Exchange (ONNX) is an open standard for machine learning interoperability. From 1.2.1 until 1.22.0, ConvTranspose shape inference processes a weight tensor with fewer than three dimensions without a kernel_shape attribute by deriving an empty effective_kernel_shape and indexing it for the input's spatial dimensions. The flaw affects convTransposeShapeInference_opset11() and the current inference path, and an untrusted model that can pass onnx.checker.check_model() can crash onnx.shape_inference.infer_shapes(), causing denial of service without demonstrated confidentiality or integrity impact. This issue is fixed in version 1.22.0.

Open Neural Network Exchange (ONNX) is an open standard for machine learning interoperability. Prior to 1.22.0, RNNShapeInference_opset1_to_6() processes malformed RNN and LSTM nodes from opsets 1 through 6 and GRU nodes from opsets 3 through 6 without verifying that the first input has rank three. The function accesses dim(0) and dim(1) on an undersized protobuf shape, which can read outside the intended repeated field and terminate an application when onnx.shape_inference.infer_shapes() processes an untrusted model that can pass onnx.checker.check_model(). This issue is fixed in version 1.22.0.

vLLM is an inference and serving engine for large language models. From 0.16.0 until 0.32.0, the /v1/responses endpoint allows an authenticated caller to place an attacker-controlled template in chat_template_kwargs[chat_template] when no server template is configured. ChatParams.get_apply_chat_template_kwargs() preserves that value without enforcing trust_request_chat_template, so the request can bypass the intended trust policy and cause CPU or memory exhaustion through Jinja rendering. This issue is fixed in version 0.32.0.

A missing authorization flaw was found in Kiali, where the namespace health, ztunnel pod log, and ztunnel configuration dump API endpoints returned data about namespaces the caller had no RBAC access to, reading ztunnel data with the Kiali ServiceAccount without checking the caller's permissions. Additionally, the openshift authentication strategy accepted tokens that could not view any namespace. An attacker with any valid OpenShift token could read other tenants' workload names, health, traffic topology, workload identities, pod addresses and mesh policies, resulting in information disclosure across tenant boundaries.

vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the OpenAI-compatible server configures Starlette CORSMiddleware with allowed_origins set to a wildcard by default. A malicious web page can use a victim's browser to reach an unauthenticated vLLM server on localhost, an internal network, or a VPN, call endpoints including GET /v1/models and POST /v1/chat/completions, consume GPU resources, and read responses. Cookie-based cross-site request forgery does not apply because credentials are disabled by default, and deployments requiring bearer authentication are exposed only when the caller already has a usable token. This issue is fixed in version 0.32.0.

nforwardauth is a forward auth service that allows the use of a single authentication middleware for multiple sites. Versions prior to 1.5.1 accept an unvalidated `r` query parameter on the `/login` endpoint and use them as the post-authentication redirect destination. An unauthenticated attacker can create a legitimate `nforwardauth` login URL that redirects a victim to an attacker-controlled website after the victim successfully authenticates. This issue may facilitate phishing and other social-engineering attacks by abusing the trust associated with the legitimate authentication domain. This vulnerability does not provide an unauthenticated ForwardAuth bypass. Version 1.5.1 contains a patch.

vLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes max_token_id with an off-by-one upper bound, allowing a pre-tokenized prompt ID equal to the vocabulary size to pass validation. For a tensor-parallel size of one and an exact multiple-of-64 vocabulary, VocabParallelEmbedding.forward does not mask that value before F.embedding uses it, causing a CUDA device-side assertion and terminating the engine. This issue is fixed in version 0.32.0.

vLLM is an inference and serving engine for large language models. Prior to 0.32.0, ImageMediaIO passes attacker-controlled image bytes to Pillow through Image.open(BytesIO(data)) without restricting decoder selection, and the MIME label does not constrain the selected format. Image-capable request paths through parse_image() and fetch_image() can therefore dispatch an EPS payload to host components. A host with Ghostscript affected by CVE-2024-29510 can execute commands with the vLLM process's privileges, while environments using Pillow 12.0.0 through 12.2.0 can trigger CVE-2026-59203 and exhaust media-loading workers. Deployments without the applicable vulnerable dependency are not affected by that corresponding impact. This issue is fixed in version 0.32.0.

A security vulnerability has been detected in Bdtask Garage 24 Garage Management Software 1.0. This issue affects some unknown processing of the file /edit_profile of the component Profile. The manipulation of the argument first_name/last_name leads to cross site scripting. The attack may be initiated remotely. The exploit has been disclosed publicly and may be used. The vendor was contacted early about this disclosure but did not respond in any way.

Burrito is a Terraform Automation Collaboration Software (TACoS) Kubernetes operator. Prior to 0.13.0, `POST /api/webhook` accepts unsigned GitHub and GitLab events when a `TerraformRepository`'s credentials omit `webhookSecret` or set it to an empty value, including credentials configured for git authentication and polling. `Credential.WebhookSecret` is optional, and `go-playground/webhooks` skips signature or token verification when the secret is empty. Anyone who can reach `burrito-server`'s endpoint and knows a tracked repository's URL can forge events to create or delete `TerraformPullRequest` custom resources with attacker-chosen `Branch`, `Base`, and `ID` fields. For GitHub `pull_request` events, the controller fetches changed files using the real pull request number but does not cross-check the payload's branch against that pull request. Pairing a real, existing pull request number with an unrelated existing branch in the same repository can therefore cause an unauthorized `terraform plan` for affected layers using the operator's stored git and cloud credentials. Code analysis also indicates that the resulting plan can be posted to the real pull request's comment thread, exposing and misattributing another branch's infrastructure changes. Auto-apply is disabled for these temporary layers, so this vulnerability alone cannot force a `terraform apply`; plan-time provider and data-source code execution depends on the selected branch's contents. This issue is fixed in version 0.13.0.

vLLM is an inference and serving engine for large language models. From 0.11.1 until 0.32.0, the DeepSeek-OCR, DeepSeek-OCR-2, and Unlimited-OCR image path allows a remote client that can reach the OpenAI-compatible API to submit a legal image whose normalized pixels sum to zero when that image is alone in its per-step encoder batch. _parse_and_validate_image_input returns no embedding, embed_multimodal propagates that value, and sanity_check_mm_encoder_outputs triggers a fatal assertion that shuts down the engine for all users. This issue is fixed in version 0.32.0.

When parsing an AS range in Routinator, if the maximum value of the range is smaller than the minimum value of the range, Routinator will never reach the maximum value and run out of memory.

Routinator runs out of open connections on the HTTP server when many connections are opened but never closed, leaving the HTTP server unavailable. This condition can be triggered maliciously by an attacker by opening a large number of connections to the HTTP server. This only affects users that make their HTTP or RTR server available to untrusted networks.

Routinator exits when an rsync URI for a file obtained over RRDP contains a path that cannot be stored on disk, such as too long file names, and folder names that overlap with existing file names.

Routinator stores a newly-retrieved TA certificate via HTTPS in cache without validation, potentially overwriting a known good TA certificate even if validation fails.

When Routinator encounters a CRL with many entries, it may hang or run out of memory. If Routinator encounters a CRL file with the wrong encoding, it may exit.

When Routinator encounters an RRDP notification file with gaps in the delta entries, it is still accepted and parsed, potentially leading to missing data. If it encounters too many delta entries it may run out of memory.

Routinator runs out of memory when the JSON output is requested many times concurrently. This condition can be triggered maliciously by an attacker by opening a large number of connections to the HTTP server. Additionally Routinator runs into a stack overflow and exits when a payload item with too much metadata is discarded at once. This only affects users that make their HTTP or RTR server available to untrusted networks.

A weakness has been identified in Bdtask Garage 24 Garage Management Software 1.0. This vulnerability affects unknown code of the component Backend. Executing a manipulation can lead to improper authorization. The attack can be launched remotely. The exploit has been made available to the public and could be used for attacks. The vendor was contacted early about this disclosure but did not respond in any way.

vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the Phi-4 multimodal audio encoder in phi4mm_audio.py and phi4mm_utils.py uses eager attention by default, allowing a remote client to submit one legal long audio clip that causes MultiHeadedAttention.forward_attention to materialize and retain full attention matrices in self.attn. The scaled dot-product attention branch does not materialize these matrices. The resulting quadratic GPU-memory use can cause an out-of-memory condition and terminate EngineCore for all users. This issue is fixed in version 0.32.0.

vLLM is an inference and serving engine for large language models. Prior to 0.32.0, an authenticated caller can place forged shared-memory address and monotonic_id values in public multimodal features.kwargs_data. In scale-out deployments with shared-memory multimodal caching enabled, workers trust those values and pass attacker-selected bytes to pickle.loads(), allowing arbitrary Python execution even when insecure serialization is disabled. This issue is fixed in version 0.32.0.

A flaw was found in GStreamer Editing Services (gst-editing-services). A GES project file can embed an arbitrary GStreamer pipeline description as an effect definition. When such a file is loaded, either explicitly by an editing application or automatically by a media player or file-manager thumbnailing service that uses GStreamer autoplugging, the pipeline description is parsed and instantiated without restriction on which elements may be used or which properties may be set. A remote attacker who can deliver a crafted project file to a victim's system can trigger reading or writing of arbitrary local files, issue outbound network requests to attacker-controlled servers, or cause the application to crash. Under certain conditions the file-write capability may be leveraged to achieve code execution.

vLLM is an inference and serving engine for large language models. From 0.19.0 until 0.32.0, authenticated callers can supply a request chat template when no server-side --chat-template is configured, or a malicious model repository can provide a template that is compiled and rendered by tokenizer.apply_chat_template() without a wall-clock timeout, output-size limit, or iteration bound. Rendering runs in a ThreadPoolExecutor with one worker by default, so a nonterminating render can block /v1/responses and other chat-template users after the request ends, while an output-expanding render can exhaust server memory. Harmony models are not affected because they do not use this Jinja path. This issue is fixed in version 0.32.0.

Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in bPlugins LightBox Block lightbox-block allows Stored XSS.This issue affects LightBox Block: from n/a through 1.2.0.

A weakness has been identified in OctoPrint 1.0.0. This impacts the function getSettings of the file src/octoprint/server/ajax/settings.py of the component API Key Handler. This manipulation causes information disclosure. The attack is possible to be carried out remotely. The exploit has been made available to the public and could be used for attacks. Upgrading to version 1.1.0-rc1 will fix this issue. Patch name: eccc9d6fbde0c46cdd6e255fd5ce9a30077ba954. You should upgrade the affected component. Affected is a 2013 release where access control was not yet implemented and the endpoint was part of first-run setup). The API key has been admin-gated since 1.2.0 and GET /api/settings is protected by Permissions.SETTINGS_READ plus reauthentication in all current releases.

Missing Authorization vulnerability in ThimPress LearnPress learnpress allows Exploiting Incorrectly Configured Access Control Security Levels.This issue affects LearnPress: from n/a through 4.4.11.

Python Social Auth is a social authentication and registration mechanism. Prior to 6.2.0, OpenIdConnectAuth in social_core/backends/open_id_connect.py uses get_and_store_nonce to place the OpenID Connect nonce in a shared OpenIdConnectAssociation table, and get_nonce and validate_claims verify only that the nonce exists and has not expired rather than binding it to the browser session and OAuth state that initiated the login. An attacker who obtains a victim's unused authorization code can start a separate login to obtain valid state for the attacker's browser, submit the victim's code with that state, and pass nonce validation because the victim nonce is globally available. With DEFAULT_USE_PKCE disabled in the default OpenIdConnectAuth configuration, the attacker can be logged in as the victim. This issue is fixed in version 6.2.0.