ClawHub
Back to skill

Security audit

Dlazy Kling V3

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed dLazy video-generation wrapper, with the main cautions being third-party API use, local API-key storage, and uploads of files you pass to it.

Install only if you intend to use dLazy/Kling for video generation. Treat prompts and any local media paths as data sent to dLazy, monitor credit usage, and protect or revoke the API key stored in ~/.dlazy/config.json if needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger keywords are broad everyday phrases such as '生成视频' and related generic wording, which can cause the skill to activate for ordinary user requests that are not explicitly asking to invoke this external tool. Because this skill sends prompts and potentially local file paths or media to remote SaaS endpoints, accidental activation can lead to unintended data disclosure, cost-incurring API calls, or execution of an external CLI without sufficiently specific user intent.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger keywords include generic phrases like "generate video" and "text to video, image to video," which are broad enough to match many ordinary user requests unrelated to this specific provider. In an agent ecosystem, this can cause the skill to be auto-selected unexpectedly, leading users to invoke a third-party hosted API, upload local media, or be prompted for authentication when they did not explicitly choose this integration.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.