ClawHub
Back to skill

Security audit

Food Production Sales Operation

Security checks across malware telemetry and agentic risk

Overview

This skill is a food e-commerce operations playbook with no executable code, installer, persistence, credential use, or hidden data handling.

This appears safe to install as an advisory playbook, but users may want to invoke it manually or narrow its trigger phrases; only grant file or terminal access if their agent platform requires it, since the skill text itself does not need those permissions.

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 (1)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger list contains several broad, generic phrases such as '运营体系', '工作体系', and '电商复盘' that can match ordinary business discussions unrelated to this specific skill. This can cause unintended invocation, exposing users to irrelevant authoritative guidance and increasing the chance that the agent applies domain-specific operational instructions in the wrong context.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.