Container

A critical vulnerability (CVE-2026-53488) exists in the containerd CRI plugin where image configuration `LABEL` instructions are propagated to containers without validation, allowing an attacker to inject and execute arbitrary commands with host-root privileges on the underlying host when a maliciously crafted container image is pulled and processed by specific plugins.

A high-severity vulnerability (CVE-2026-53489) in containerd's CRI plugin allows an unprivileged attacker to read arbitrary files on the host system by crafting a malicious checkpoint with a symlink that `containerd` follows during `container.log` restoration, enabling data exfiltration via `kubectl logs`.

An attacker can achieve unauthenticated remote code execution (RCE) in Crawl4AI Docker deployments by injecting malicious Chromium launch arguments, such as `--utility-cmd-prefix` and `--no-zygote`, into the `browser_config.extra_args` field of the API request, allowing for arbitrary command execution as the container's runtime user.

A high-severity authentication bypass vulnerability in PraisonAI versions prior to 4.6.61 allows unauthenticated attackers to invoke any registered agent by setting the `PRAISONAI_CALL_AUTH=disabled` environment variable, potentially leading to arbitrary code execution or system compromise.

This rule detects Linux process executions that reference /etc/kubernetes/manifests in process arguments, which may indicate tampering with static pod manifests for persistence or privilege escalation in Kubernetes environments.

A vulnerability exists in Docker where a malicious container image can execute arbitrary code with host root privileges by exploiting the decompression of compressed archives uploaded via the `PUT /containers/{id}/archive` endpoint, tracked as CVE-2026-41567.

Detects execution of curl or wget from processes running inside OCI/runc-backed containers, potentially indicating ingress tool transfer or data exfiltration after a container breakout.

The rule detects the use of the 'kubectl get secrets --all-namespaces' command, which enumerates secret resources across the entire Kubernetes cluster, potentially aiding credential discovery, privilege escalation, or lateral movement by attackers.

Portainer versions 2.33.0 through 2.33.7, 2.39.0 through 2.39.1, and 2.40.0 through 2.40.9 are vulnerable to CVE-2026-44850, a bind-mount restriction bypass via the `HostConfig.Mounts` array allowing regular users to mount host paths into containers and potentially compromise the host filesystem.

This brief outlines how Linux cgroups, a kernel feature for resource management, can be repurposed to provide valuable telemetry for detecting malicious processes, particularly in systemd, Docker, and Kubernetes environments, aiding in investigations of server compromises.

A local attacker can exploit a vulnerability in Podman HyperV Machine to execute arbitrary program code with administrator privileges, leading to complete system compromise.

A vulnerability in the distribution toolkit prior to 3.1.0 allows a malicious upstream registry or man-in-the-middle attacker to redirect authentication requests, potentially exposing upstream credentials.

The rule detects the creation or modification of an authorized_keys file inside a container, a technique used by adversaries to maintain persistence on a victim host by adding their own public key(s) to enable unauthorized SSH access for lateral movement or privilege escalation.

An attacker may deploy a pod within the kube-system namespace in Kubernetes to mimic legitimate system pods and evade detection.

This rule detects the execution of curl or wget from within runc-backed containers on Linux systems monitored by Auditd Manager, indicating potential ingress tool transfer or data exfiltration by attackers who have compromised the container.

Detection of runc init child processes with root effective user and non-root login user ID, indicating potential container privilege escalation.

The rule detects execution of container runtime CLI tools (ctr, crictl, nerdctl) with arguments indicating container creation, command execution inside existing containers, image manipulation, or host filesystem mounting, potentially leading to container escape and privilege escalation.

This rule detects nsenter executions that target a PID with a namespace target flag, a common pattern used to attach to the host init namespace from a container or session and run with host context, potentially escalating privileges.

The rule detects network connection attempts to the Kubernetes Kubelet API ports 10250 and 10255 on internal IP ranges from Linux hosts, indicating potential lateral movement within container and cluster environments.

Detects execution of container runtime CLI tools (ctr, crictl, nerdctl) with arguments indicating container creation, command execution inside existing containers, image manipulation, or host filesystem mounting, potentially leading to privileged container creation and unauthorized access to sensitive data.

An unusual process connecting to a container runtime Unix socket like Docker or Containerd can indicate an attacker attempting to bypass Kubernetes security measures for container manipulation.

Detection of potential direct Kubelet access via process arguments in Linux containers, which could lead to enumeration, execution, or lateral movement within the Kubernetes cluster.

This rule detects the execution of kubeletctl, a command-line tool used to interact with the Kubelet API, on Linux hosts, potentially leading to discovery and lateral movement within Kubernetes environments.

The rule detects nsenter executions from inside a monitored Linux container that include a namespace target flag (-t or --target), which can be abused to escape container isolation.

This rule detects the execution of kubeletctl inside a container, which can be used to enumerate the Kubelet API or other resources inside the container, potentially indicating lateral movement attempts within the pod.

Gotenberg version 8.30.1 and earlier is vulnerable to argument injection, where an unauthenticated attacker can inject arbitrary ExifTool pseudo-tags via newline characters in metadata values, leading to arbitrary file manipulation within the container filesystem.