Cloudcraft

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate Cloudcraft integration, but it exposes broad authenticated API access that can modify or delete Cloudcraft data without clear built-in confirmation rules.

Install only if you trust Membrane and intend to let an agent access Cloudcraft. Use a least-privileged Cloudcraft/Membrane connection, prefer read-only discovery and prebuilt actions, and require the agent to show and confirm any POST, PUT, PATCH, or DELETE request before it runs.

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: 86% confidence
Finding: The skill explicitly instructs the agent to send direct proxy requests to the Cloudcraft API and lists support for mutating HTTP methods like POST, PUT, PATCH, and DELETE, but it does not require confirmation, scope checks, or warn about destructive operations. In an agent setting, this increases the chance of unintended data modification, overbroad requests, or sensitive data transmission through raw endpoints outside safer prebuilt actions.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal