ClawHub
Back to skill

Security audit

Workflow Dlc Agent Scenario

Security checks across malware telemetry and agentic risk

Overview

This skill is a structured product-design workflow for AI agent scenarios and does not contain executable code, hidden data collection, or privileged actions.

Safe to install for agent product-design work. Treat the PageContext and log examples as design templates, avoid including secrets, personal data, or sensitive business plans in generated logs, and review the referenced external lesson file if it exists in your environment.

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
84% confidence
Finding
The manifest uses broad trigger phrases such as '设计 Agent' and '做 AI 助手', which can cause the skill to activate for a wide range of generic product or assistant-design requests. This creates an overbroad routing risk where the skill may run in unintended contexts, potentially steering conversations, collecting context, or logging data when the user did not intend to invoke this workflow.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly instructs the agent to read page context automatically and later defines structured interaction logging, but it does not include any user-facing notice, consent boundary, or data-minimization rule. In an embedded product setting, this can expose sensitive operational metadata, selected records, form contents, locale, or market information to the agent and persistent logs without clear user awareness.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.