Cloudentity

• Outcome: Review
• Latest audit: May 29, 2026 at 9:37 PM UTC
• Version:
  1.0.2

Overview

This Cloudentity skill is coherent, but it gives an agent broad authenticated control over a security platform without enough safeguards around write or delete actions.

Install only if you trust Membrane and need agent access to Cloudentity. Use a least-privileged Cloudentity account, prefer discovered Membrane actions over raw proxy requests, require explicit confirmation before any create, update, delete, policy, client, scope, gateway, or entitlement change, and audit or revoke the connection after use.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill explicitly documents a raw request proxy that supports destructive HTTP methods like PUT, PATCH, and DELETE without requiring confirmation, scoping guidance, or warning about production impact. In an agent context, this increases the chance that a model will perform high-impact state-changing API operations directly against Cloudentity resources based on ambiguous prompts or mistaken assumptions.

VirusTotal

65/65 vendors flagged this skill as clean.