Alicloud Ai Entry Modelstudio

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Alibaba Cloud Model Studio router with expected SDK, credential, API-call, and local-output behavior.

Install this only if you intend to use Alibaba Cloud Model Studio services. Review the downstream skill selected for each task, use a limited DashScope API key, be explicit before mutating operations, consider pinning the dashscope package version, and clear saved output files if they contain sensitive content.

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 (1)

Intent-Code Divergence

Medium
Confidence
76% confidence
Finding
The skill is presented as a routing entry point, but its workflow instructs the agent to confirm identifiers, run a read-only query, and execute target operations. That scope expansion can cause an agent to perform real API actions from a skill the user may invoke expecting only capability selection, increasing the risk of unintended external calls or mutating operations under available credentials.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal