Google Cloud

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate Google Cloud connector, but it grants broad cloud access and direct API capability without enough guardrails for changes or deletes.

Install only if you are comfortable granting Membrane and your agent access to a Google Cloud environment. Use a dedicated least-privilege, project-scoped account, prefer read-only discovered actions, require explicit approval before any non-GET proxy request or change to IAM, secrets, networking, DNS, compute, storage, or databases, and revoke the Membrane connection when finished.

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: 93% confidence
Finding: The skill explicitly documents a generic proxy request capability supporting GET, POST, PUT, PATCH, and DELETE against Google Cloud APIs without requiring confirmation or warning about destructive operations or sensitive data transmission. In a cloud-admin context, this makes it easier for an agent to perform unintended state-changing actions, including resource deletion or configuration changes, especially if the user request is ambiguous.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal