ClawHub
Back to skill

Security audit

Alibabacloud Network Reachability Analysis

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Alibaba Cloud network diagnostics skill, but users should handle cloud credentials carefully and understand that it creates temporary NIS analysis tasks.

Install only if you want an agent to run Alibaba Cloud NIS reachability checks and CloudMonitor read queries. Use a least-privilege RAM user or temporary credentials, avoid pasting real access keys into chat or command examples, and review any Aliyun CLI installation or auto-plugin step before running it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The skill repeatedly characterizes the workflow as "read-only" and later states that no cloud resources are created or modified, yet it explicitly instructs the agent to call `CreateAndAnalyzeNetworkPath`. Even if the API only creates temporary analysis tasks rather than business infrastructure, this is still a state-changing operation and the mismatch can mislead users, bypass stricter approval flows, or violate environments that permit only truly non-mutating actions.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The guide explicitly recommends exporting long-lived Alibaba Cloud access keys as environment variables, but it does not warn that such secrets can leak through inherited child processes, CI job logs, crash dumps, shell startup files, or process/environment inspection in some execution contexts. In an agent-driven automation skill, this is more dangerous because credentials are likely to be propagated across subprocesses and unattended workflows, increasing exposure surface.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The guide recommends non-interactive configuration using command-line flags that include access-key secrets directly in the command invocation. Secrets passed on the command line may be captured by shell history, terminal logging, audit systems, process listings, and orchestration logs; in agent/CI automation this risk is amplified because commands are often recorded verbatim.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.