zeabur-server-ssh

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx zeabur@latest to download and execute the latest version of the vendor-provided CLI tool from the NPM registry.- [COMMAND_EXECUTION]: It provides various commands for interacting with remote servers and Kubernetes clusters, including server exec, kubectl logs, and kubectl exec.- [DYNAMIC_EXECUTION]: A fallback SSH connection method uses node -e to execute a JavaScript snippet that dynamically requires the ssh2 module and establishes a connection using provided credentials.- [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill processes output from remote server commands such as kubectl get pods, kubectl logs, and reading files like README.md.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the command outputs.
  • Capability inventory: The skill has broad capabilities including executing arbitrary shell commands on remote servers, accessing Kubernetes logs, and executing processes inside containers.
  • Sanitization: There is no evident sanitization of the remote command output before it is returned to the agent context.- [CREDENTIALS_UNSAFE]: In the manual SSH fallback path, the skill retrieves a server password via the CLI and interpolates it directly into a shell command (sshpass) or a Node.js script string. This practice can lead to credential exposure in system logs or process managers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 04:29 PM