ClawHub
Back to skill

Security audit

矢量图生成 Recraft V4 Pro Vector

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed wrapper around the dLazy hosted vector-generation CLI and does not show hidden or purpose-mismatched behavior.

Before installing, verify you trust the dLazy CLI package and understand that prompts, selected input files, and generated outputs go through dLazy's hosted service; use npx if you want to avoid a persistent global install, and rotate the API key if needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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)

Natural-Language Policy Violations

Medium
Confidence
92% confidence
Finding
The skill embeds agent-directed instructions in Chinese and explicitly tells the agent what to say and do on certain failures, without any indication that the end user opted into Chinese-language responses. This can override user expectations, reduce transparency, and create phishing-like trust issues by steering the agent into fixed messaging and links regardless of user language preference.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.