ClawHub
Back to skill

Security audit

酷家乐AI室内设计

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Kujiale interior-design workflow that uses a local token, image upload, script execution, and account quota in ways that fit its stated purpose.

Use this only when you intend to connect a Kujiale account for interior design. Review any local Node scripts before running them, keep the .kjlconfig.json token private, upload only floor-plan images you want sent to Kujiale, and confirm quota-consuming layout actions before proceeding.

SkillSpector

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 (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
80% confidence
Finding

The activation trigger is overly broad: any request mentioning interior design or renovation design may invoke a skill that performs command execution, file monitoring, network calls, and potentially billable operations. In an agent environment, underspecified routing increases the chance of unintentional invocation, causing privacy-impacting local access or unintended external actions without sufficiently explicit user intent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill advertises local command execution, reading credential material, and monitoring a local inbound media directory, but does not present a prominent upfront warning or consent step before these behaviors. In agent contexts, that is dangerous because users may think they are invoking a simple design assistant while the skill is authorized to inspect local paths, consume secrets, and perform external uploads on their behalf.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.