ClawHub
Back to skill

Security audit

Kujiale 3D Model Upload

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Kujiale model-upload helper that uses user-provided API credentials and does not show hidden or unrelated behavior.

Use --dry-run first. Only run the real flow in a trusted shell because it can upload and submit a model to your Kujiale tenant, review the default location and brand category values, and avoid sharing terminal or CI logs because the app key is printed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The code prints `appKey` to the console during real runs, which can expose credentials in terminal history, CI logs, agent transcripts, or shared observability systems. While the app secret is not printed here, the finding is still valid because credential identifiers should be treated as sensitive and the skill context involves automated execution where logs may be retained or visible to others.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.