Wealth Copilot Digital Employee

Security checks across malware telemetry and agentic risk

Overview

This is a non-executable wealth-management reference skill with a disclosure inconsistency around a listed external Qieman endpoint, but no code, tools, persistence, or credential handling.

Install only if you want an advisory wealth-management reference framework. Be aware that the documentation names an external Qieman/DashScope endpoint despite claiming no network calls; do not connect it to live accounts, client data, or investment workflows unless you separately review the data flow, compliance requirements, and human approval process.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The skill explicitly claims there are no network calls, yet it documents an external service endpoint for Qieman. This inconsistency can mislead reviewers and downstream systems about the skill's actual trust boundary, causing unsafe approval or deployment assumptions around outbound connectivity and third-party data handling.

Description-Behavior Mismatch

Medium
Confidence
88% confidence
Finding
The manifest frames the skill as educational, advisory-only, and non-executable, but the body describes concrete financial-service tooling such as fund diagnosis, portfolio backtesting, asset-allocation planning, and chart rendering. That mismatch can cause operators to under-scope risk, because a skill presented as passive reference material may in practice facilitate higher-risk financial analysis workflows or external-tool integration.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal