Daily Digest
User Safety: safe
The most urgent item is CVE-2026-73842 in OpenChoreo, a Kubernetes-native developer platform, which exposes critical internal APIs without authentication, allowing unauthenticated remote attackers to read secrets, mutate workloads, and execute commands across connected clusters. This is compounded by active exploitation reports and a CVSS score of 9.0 (Critical).
Summary: A race condition in the OpenChoreo Kubernetes platform exposes internal APIs (`/api/proxy/`, `/api/exec/`, `/api/wirelogs/`) without requiring client certificates or tokens, enabling unauthenticated network-reachable attackers to read tenant Kubernetes Secrets, mutate workloads, and execute commands across connected data planes. No authentication is required, and the vulnerability is exploitable over the network.
Affected Features / Interoperability: Impacts OpenChoreo versions prior to 1.0.3, 1.1.3, and 1.2.0-rc.2. Affects internal cluster gateway services and any connected data planes relying on OpenChoreo for workload management. Downstream impacts include unauthorized access to Kubernetes Secrets, potential compromise of managed clusters, and lateral movement within Kubernetes environments.
Source:
Actions:
Summary: OpenChoreo Workflow Plane templates interpolate developer-controlled parameters into shell commands executed via `sh -c`, allowing authenticated users to inject arbitrary commands into workflow pods. Affected privileged Podman templates also lack `hostUsers: false`, increasing exposure.
Affected Features / Interoperability: Impacts OpenChoreo versions prior to 1.0.4, 1.1.4, and 1.2.0-rc.2. Affects workflow execution pipelines and any CI/CD integrations using OpenChoreo templates. Could lead to command injection in workflow pods and potential privilege escalation.
Source:
Actions:
Summary: The cluster-proxy service-proxy component in Red Hat Advanced Cluster Management (RHACM) and multicluster-engine (MCE) appends impersonation group headers to proxied requests without removing caller-supplied values. An authenticated hub principal can inject an `Impersonate-Group` header to escalate privileges to cluster-admin on all managed clusters.
Affected Features / Interoperability: Impacts RHACM and MCE deployments using the service-proxy component. Affects IAM policy enforcement, cluster management, and cross-cluster operations. Could result in full administrative control over managed clusters.
Source:
Actions:
Summary: Improper validation of Kubernetes client certificate renewal in Open Cluster Management (OCM) allows a managed cluster administrator to forge a client certificate approved by the OCM controller, enabling cross-cluster privilege escalation and potential control over other managed clusters, including the hub cluster.
Affected Features / Interoperability: Impacts OCM-based systems including RHACM. Affects certificate-based authentication, cluster registration, and inter-cluster communication. Could lead to unauthorized access across multiple clusters.
Source:
Actions:
Summary: Red Hat Advanced Cluster Security for Kubernetes (RHACS) Central does not limit the depth of GraphQL queries on its authenticated API, allowing authenticated users with valid API tokens to send deeply nested queries that consume excessive resources, causing denial of service for the management plane.
Affected Features / Interoperability: Impacts RHACS Central management plane. Affects security monitoring, policy enforcement, and cluster visibility dashboards. Could disrupt security operations and incident response capabilities.
Source:
Actions:
Summary: A race condition in Microsoft Azure Attestation service and Device Health Attestation Service allows an unauthorized attacker to execute code over a network without authentication.
Affected Features / Interoperability: Impacts Azure Attestation and Device Health Attestation services. Affects device identity verification, attestation workflows, and integration with Azure Active Directory. Could lead to unauthorized code execution in attestation pipelines.
Source:
Actions:
Summary: An integer overflow in Microsoft Azure Attestation service and Device Health Attestation Service allows an unauthorized attacker to execute code over a network without authentication.
Affected Features / Interoperability: Impacts Azure Attestation and Device Health Attestation services. Affects device identity verification, attestation workflows, and integration with Azure Active Directory. Could lead to unauthorized code execution in attestation pipelines.
Source:
Actions:
Summary: A SQL injection vulnerability in a legacy dashboard widget API in Google Cloud Google SecOps (Chronicle SOAR) allows an authenticated attacker to execute blind SQL queries using crafted request parameters. The vulnerability was patched in version 6.3.85, and no customer action is needed.
Affected Features / Interoperability: Impacts Google Cloud Google SecOps (Chronicle SOAR) versions prior to 6.3.85. Affects dashboard widgets, reporting features, and integration with security orchestration tools. Could lead to unauthorized data access or manipulation.
Source:
Actions:
Summary: The `/create_project` endpoint in UpTrain (version 0.7.1 and prior) is vulnerable to remote code execution via the `checks` and `metadata` parameters. Any authenticated user with access to UpTrain can execute arbitrary code in the context of the host running UpTrain, typically within a Docker container. No known patch is available.
Affected Features / Interoperability: Impacts UpTrain versions 0.7.1 and prior. Affects project creation workflows, AI application evaluation pipelines, and Docker-based deployments. Could lead to full container compromise and lateral movement.
Source:
Actions:
Summary: The `/add_prompts` endpoint in UpTrain (version 0.7.1 and prior) is vulnerable to remote code execution via the `checks` and `metadata` parameters. Any authenticated user with access to UpTrain can execute arbitrary code in the context of the host running UpTrain, typically within a Docker container. No known patch is available.
Affected Features / Interoperability: Impacts UpTrain versions 0.7.1 and prior. Affects prompt management workflows, AI application evaluation pipelines, and Docker-based deployments. Could lead to full container compromise and lateral movement.
Source:
Actions:
Summary: The `/new_run` endpoint in UpTrain (version 0.7.1 and prior) is vulnerable to remote code execution via the `checks` and `metadata` parameters. Any authenticated user with access to UpTrain can execute arbitrary code in the context of the host running UpTrain, typically within a Docker container. No known patch is available.
Affected Features / Interoperability: Impacts UpTrain versions 0.7.1 and prior. Affects run execution workflows, AI application evaluation pipelines, and Docker-based deployments. Could lead to full container compromise and lateral movement.
Source:
Actions: