ClawHub
Back to skill

Security audit

智能菜谱

Security checks across malware telemetry and agentic risk

Overview

This is a simple recipe helper that directs users to a third-party recipe website and does not install code or access local data.

Install this only if you are comfortable with the agent suggesting a third-party recipe website for some cooking prompts. For normal recipe advice, you can ask the agent to answer directly, and avoid entering sensitive personal information into the external site.

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
89% confidence
Finding
The skill description includes broad activation language such as recipe generation, cooking ideas, 菜谱推荐, and 吃什么, which are common everyday requests and not uniquely scoped to this tool. In an agent environment, this can cause the skill to activate for many general cooking queries and unnecessarily steer users to an external website, increasing the chance of inappropriate routing and unsolicited third-party exposure.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The listed trigger keywords include highly ambiguous phrases like '今天吃什么' and '用这些食材能做什么,' which are common requests unrelated to this specific service. Because the skill primarily redirects users to https://eat.lz-t.top/, overbroad triggers can hijack normal assistant interactions and funnel users to an external domain without clear necessity or informed user intent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.