ClawHub
Back to skill

Security audit

kuaidi100-user

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate shipping assistant, with disclosed handling of login, address, order, and browser-opening flows.

Install only if you trust the kuaidi100 CLI package and are comfortable letting it handle shipping addresses, order details, and an SSO browser login. Before using it, confirm any browser-opened login or order link looks like the expected kuaidi100 flow, and do not allow order creation or cancellation unless you have reviewed the details.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
84% confidence
Finding
The skill's trigger description is overly broad, covering common shipping, pricing, and tracking phrases that may appear in ordinary conversation. This can cause the agent to invoke a logistics skill in unintended contexts, increasing the chance of unnecessary authentication prompts, exposure of order-management capabilities, or accidental initiation of sensitive shipping workflows involving personal address data.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The workflow directs use of `kd100 auth login --open`, which causes an external browser to launch automatically, but the user warning comes after the command is executed. Launching an external application without prior notice or consent can surprise users and may facilitate phishing or trust-boundary confusion if the CLI or returned URL is compromised.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The workflow permits automatically opening `data.url` in a browser after order creation without requiring prior user confirmation. Even though link and QR display are mandatory first, auto-opening a returned URL expands the attack surface if the URL is malformed, unexpected, or influenced by an upstream compromise, and it violates the principle of explicit user consent for launching external applications.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.