SZZG007 Customer CRM

• Outcome: Review
• Latest audit: May 29, 2026 at 9:33 PM UTC
• Version:
  1.0.0

Overview

This CRM skill has a legitimate purpose, but it asks the agent to handle sensitive customer records, cloud sync, credentials, message-derived updates, and automated cleanup without enough boundaries or confirmations.

Review this before installing if your CRM contains personal, regulated, or commercially sensitive data. Use it only with explicit user approval for cloud sync, narrow service-account permissions, protected credential storage, backups, dry-run previews for cleanup/conflict resolution, and clear limits on which email/message sources and customer fields the agent may read or change.

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 (2)

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill explicitly supports syncing CRM/customer data to Feishu Bitable and Google Sheets, which are external services, but it does not present a clear user-facing warning, consent step, or data handling limits before transfer. Because the data includes personal and business contact details, silent or default cloud synchronization can cause unintended disclosure, cross-border transfer, or compliance issues.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill states it can automatically detect customer status updates from emails/messages, which implies monitoring and processing communication content that may contain sensitive personal, commercial, or confidential information. Without a clear notice, consent model, and scope limitation, users may unknowingly enable surveillance-like processing and over-collection of message data.

VirusTotal

65/65 vendors flagged this skill as clean.