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 versions from 0.16.0 up to but not including 0.32.0 contain a vulnerability in the /v1/responses endpoint that allows an authenticated user to supply a malicious template in chat_template_kwargs[chat_template]. This bypasses the intended trust policy and can lead to uncontrolled CPU or memory consumption during Jinja template rendering. The issue is resolved in version 0.32.0.
vLLM versions prior to 0.32.0 have a permissive cross-origin resource sharing (CORS) policy that allows any origin by default. This misconfiguration enables a malicious web page to interact with an unauthenticated vLLM server running on localhost, internal networks, or VPNs, potentially consuming GPU resources and reading responses. The 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 versions prior to 0.32.0 contain an off-by-one error in token ID validation that can cause a CUDA device-side assertion failure and engine termination when processing certain pre-tokenized prompts. This vulnerability 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.
CVE-2026-102649 is a critical vulnerability in padok-team's burrito Kubernetes operator prior to version 0.13.0. It allows unauthenticated attackers to forge GitHub and GitLab webhook events if the webhookSecret is omitted or empty in TerraformRepository credentials. This can lead to unauthorized creation or deletion of TerraformPullRequest resources with attacker-controlled parameters. The flaw enables unauthorized terraform plan operations on affected layers and disclosure of plan results in pull request comments. Auto-apply is disabled, so forced terraform apply is not possible. The issue is fixed in burrito version 0.13.0.
vLLM versions from 0.11.1 up to but not including 0.32.0 contain a reachable assertion vulnerability (CWE-617) in the image processing path for certain OCR models. A remote client with access to the OpenAI-compatible API can submit a specific image that causes a fatal assertion failure, resulting in a shutdown of the inference engine for all users. This denial-of-service condition 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.
CVE-2026-116384 is an improper authorization vulnerability in Bdtask Garage 24 Garage Management Software version 1.0. The flaw exists in the backend component and can be exploited remotely without user interaction. Public exploit code is available, but the vendor has not responded or issued a fix. The vulnerability has a medium severity rating with a CVSS 4.0 score of 5.3.
vLLM versions prior to 0.32.0 contain a resource allocation vulnerability in the Phi-4 multimodal audio encoder. The default use of eager attention causes large attention matrices to be fully materialized in GPU memory when processing long audio clips, leading to quadratic memory consumption. This can result in out-of-memory conditions that terminate the EngineCore process, affecting all users. The issue is resolved in version 0.32.0.