Cradl Ai

• Outcome: Review
• Latest audit: May 29, 2026 at 9:36 PM UTC
• Version:
  1.0.2

Overview

This looks like a legitimate Cradl AI integration, but it gives broad authenticated access to read, change, or delete business data without clear confirmation guardrails.

Install only if you trust Membrane and are comfortable granting access to your Cradl AI account. Use least-privilege access where possible, review any action inputs before running them, and require explicit user approval before creating, updating, bulk-changing, or deleting Cradl AI data.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• 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: 81% confidence
• Finding: The skill explicitly instructs the agent to run actions and proxy direct API requests, including methods like POST, PUT, PATCH, and DELETE, without requiring an explicit confirmation or warning before data is transmitted or modified. In an agent setting, this can lead to unintended writes, destructive actions, or disclosure of sensitive business data if the agent proceeds on ambiguous user intent.

VirusTotal

64/64 vendors flagged this skill as clean.