ClawHub
Back to skill

Security audit

DashScope LLM

Security checks across malware telemetry and agentic risk

Overview

This skill is a small, disclosed DashScope LLM CLI; the main risk is that prompts are sent to Alibaba Cloud DashScope.

Install only if you intend to use Alibaba Cloud DashScope and are comfortable sending the message text you provide to that service. Do not include passwords, API keys, personal data, or confidential business content unless that sharing is approved for your use case.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding
The skill documentation indicates the code reads `DASHSCOPE_API_KEY` from the environment, but the skill metadata shown in `SKILL.md` does not declare corresponding permissions or clearly surface that capability. Undeclared access to environment-provided secrets reduces transparency and can cause the skill to be run in contexts where operators do not realize it depends on or may expose sensitive credentials.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill sends user-supplied prompts to DashScope, a third-party API, but the description does not clearly warn users that their input leaves the local environment. This creates a data leakage risk because users may submit sensitive prompts, proprietary text, credentials, or regulated data without informed consent or appropriate handling expectations.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The script sends arbitrary user-provided text directly to a third-party cloud LLM service, but the code provides no disclosure, consent prompt, or safeguard to help users understand that their input leaves the local environment. This creates a real privacy and data-handling risk if a user enters secrets, credentials, personal data, or proprietary content during quick testing.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.