CVE-2026-11587 is a high-severity SQL injection vulnerability in Centreon affecting versions 24.10 and 25.10. It allows an authenticated user with access to ACL management to read and modify the Centreon database. The vulnerability is fixed in centreon-web versions 24.10.28 and 25.10.15.
A vulnerability in Keycloak allows a realm administrator with the "manage-realm" role to disclose information by submitting arbitrary filesystem paths as keystore parameters when creating a key provider component. This flaw enables probing of filesystem paths to determine file existence and readability by the Keycloak process, potentially aiding further attacks. The issue affects Keycloak versions prior to 26.6.4. A fix is available in version 26.6.4.
A vulnerability in org.keycloak:keycloak-services allows an authenticated user with a User-Managed Access (UMA) permission ticket for one resource to bypass per-resource access controls and gain unauthorized access to all resources of that type within the same resource server. This issue occurs when the resource server is configured in PERMISSIVE policy enforcement mode and affects typed resources with ownerManagedAccess enabled without explicit policies protecting the resource type. The vulnerability can lead to unauthorized information disclosure or modification of resources. A fix is available in version 26.6.4.
Kubeflow Pipelines enables users to build and deploy portable, scalable machine learning workflows. Prior to 2.18.0, the API server decompressed uploaded pipeline archives without adequately bounding expanded data while locating or reading the pipeline file. A small ZIP or gzip-compressed tar archive can consume disproportionately large memory and exhaust API-server resources. Multipart pipeline and pipeline-version upload processing performs decompression before namespace authorization, so a reachable caller does not need a successful pipeline-create authorization decision. URL-import handlers authorize before fetching and decompressing a remote archive. The amount of expansion required to interrupt service depends on deployment resource limits and concurrent activity, and no specific archive is guaranteed to crash every deployment. This issue is fixed in version 2.18.0.
A vulnerability in Keycloak Policy Enforcer allows any authenticated user to bypass all authorization policies, including role, scope, and UMA permission checks. This occurs when an attacker includes the configured access-denied page path within a request URL, either as a path segment or a query parameter, enabling unauthorized access to protected resources. The flaw affects Keycloak versions prior to 26.6.4 and 26.0.10. A fix is available in these versions.
CVE-2026-119115 is an incorrect authorization vulnerability in AWS Private CA Connector for Kubernetes (aws-privateca-issuer) before version 1.10.1. It allows an authenticated remote user to obtain certificates, including subordinate CA certificates, signed by a cluster-scoped AWS Private CA issuer that their approval policy does not authorize. The issue is fixed in version 1.10.1 and the corresponding EKS add-on version 1.10.1-eksbuild.1. Users must upgrade and ensure issuerRef.kind fields are correctly set to AWSPCAIssuer or AWSPCAClusterIssuer to match the referenced issuer.
A vulnerability in Keycloak's client registration service allows an attacker with a previously issued Registration Access Token (RAT) to re-enable clients that an administrator has disabled. This enables the attacker to reset the client's secret and potentially regain privileged API access, leading to unauthorized information disclosure and integrity compromise. The issue affects versions prior to 26.6.4 and has been assigned CVE-2026-9705 with a CVSS score of 6.5 (medium severity). A patch is available in version 26.6.4.
@eeacms/volto-chatbot version prior to 2.0.4 contains an unauthenticated proxy vulnerability that allows attackers to invoke arbitrary Danswer API operations with server-side credentials. The middleware route '**/_v1_da/**' accepts any HTTP method and path without authentication or path restrictions, forwarding requests with attached API keys or session cookies. This enables a Confused Deputy attack where attackers can perform privileged actions on the upstream Danswer API as the configured service account.
Crass versions prior to 1.0.7 have a vulnerability where parsing input containing non-ASCII characters leads to superlinear CPU consumption due to inefficient tracking of multi-byte character positions. This can be exploited by an attacker to cause excessive CPU usage and potential denial of service. The issue is fixed in version 1.0.7.
Crass versions prior to 1.0.7 contain a vulnerability where processing a stylesheet with a large number of adjacent CSS comments causes excessive recursion, leading to a SystemStackError. This occurs when the default behavior of discarding comments is used (i.e., :preserve_comments option is false). The issue is fixed in version 1.0.7, which eliminates recursion when discarding comments. As a workaround, enabling the :preserve_comments option prevents the recursion and the error.
Crass versions prior to 1.0.7 contain a vulnerability where deeply nested CSS blocks and functions can cause unbounded recursion, leading to a SystemStackError or excessive memory consumption. This issue arises because Crass lacks a depth guard when parsing CSS simple blocks and functions. The vulnerability is fixed in version 1.0.7 by introducing a configurable depth limit with a conservative default.
Kubeflow Pipelines versions prior to 2.17.0 contain an authorization bypass vulnerability in the frontend artifact routes. Authenticated users could access artifact contents from other tenants' namespaces without proper authorization checks, potentially exposing datasets, models, or pipeline outputs. This affects multi-user deployments relying on Kubeflow Pipelines for tenant isolation. The vulnerability does not allow disclosure of Kubernetes Secrets, modification of storage, or server-side code execution. The issue is fixed in version 2.17.0.
A vulnerability in Cedar JSON policy templates using the 'is in' scope condition allows malformed slot identifiers to cause linking errors or semantic mismatches. This can lead to policy panics or unintended authorization decisions, such as allowing access that should be forbidden. The issue affects only malformed JSON policies with incorrect slot identifiers in principal and resource constraints. Versions prior to 4.11.1 and 3.4.3 are affected. Patches are available in these fixed versions.
The @sylphx/pdf-reader-mcp package, when run as an HTTP server with MCP_TRANSPORT=http, fails to enforce API key authentication despite logging that it does. This allows unauthenticated access to the /mcp endpoint, enabling any client able to reach the server to invoke PDF processing tools and read any accessible PDF files on the host. The default binding to all interfaces (0.0.0.0) increases exposure risk. This vulnerability affects versions >=2.2.0 and <3.0.1 and is fixed in version 3.0.1.
PraisonAI versions prior to 4.6.78 contain a vulnerability in the JobWorkflowExecutor._exec_inline_python() function that allows remote code execution. The vulnerability arises because the sandbox intended to safely execute inline Python expressions is broken: the 'os' module is allowed to be imported, and dangerous functions like os.system() can be called. This enables an attacker to execute arbitrary OS commands with the privileges of the user running the workflow, particularly in multi-tenant or CI environments where untrusted YAML workflows may be executed.
Parse Server versions prior to 9.10.0-alpha.2 and 8.6.84 are vulnerable to a stored cross-site scripting (XSS) issue via malformed Content-Type headers in file uploads. The vulnerability arises because malformed Content-Type values that do not conform to the type/subtype format can bypass the file extension blocklist and be stored unchanged. When served from storage adapters that preserve the Content-Type header (e.g., Amazon S3, Google Cloud Storage, Azure Blob Storage), browsers may MIME-sniff the file content and execute embedded scripts, leading to stored XSS. The default GridFS adapter is not affected due to recomputing Content-Type and using X-Content-Type-Options: nosniff. Patches are available that validate Content-Type headers for well-formedness and reject malformed values.
Daptin versions up to 0.12.15 contain a vulnerability in their global LanguageMiddleware that parses the Accept-Language HTTP header without proper size or character filtering. The underlying Go language parser has quadratic-time complexity on malformed language tags using underscore separators, allowing an unauthenticated attacker to cause significant CPU resource exhaustion with a single crafted request. This can lead to denial of service by consuming excessive server CPU for extended periods.
PocketMine-MP versions prior to 5.44.2 have a vulnerability during the resource pack stage of client login. A malicious client can send multiple ResourcePackClientResponsePacket packets with a completed status, causing the server to repeatedly execute pre-spawn progression. This leads to the creation of multiple Player objects for a single session, increasing memory usage and outbound network traffic due to duplicated large pre-spawn packets. The issue was patched in version 5.44.2.
A vulnerability in SilverStripe CMS version 5 allows bypassing multi-factor authentication (MFA) during the password reset process. An attacker with access to a user's email can reset the password and gain a logged-in session without completing MFA. This issue is fixed in version 6.0.0 and does not affect CMS 6.
A vulnerability in Kubeflow Pipelines prior to version 2.18.0 allows unauthorized access to logs via the API server's log retrieval endpoint. The API server does not authorize callers against the selected run, enabling an attacker with network access and knowledge of run and node IDs to read live or archived logs. This includes the ability to read logs from unrelated pods by manipulating node IDs. The exposed logs may contain sensitive workload data or credentials. The issue is fixed in version 2.18.0.