Cubejs

Security checks across malware telemetry and agentic risk

Overview

This Cube.js skill is coherent, but it needs review because it gives an agent broad authenticated API access, including write and delete requests, without clear approval guardrails.

Install only if you trust Membrane and need agent-assisted Cube.js administration. Use a least-privilege Cube.js/Membrane connection, prefer discovered Membrane actions over raw proxy calls, and require explicit approval before any POST, PUT, PATCH, DELETE, user, role, model, or data-changing request.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill explicitly instructs the agent to send direct proxy requests to the external Cube.js API, but it does not require explicit user confirmation or warn that data may be transmitted off-platform. In an agent setting, this can lead to unintended disclosure of sensitive prompts, query contents, or business data to an external service.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal