ClawHub
Back to skill

Security audit

搜索工具免费版

Security checks across malware telemetry and agentic risk

Overview

This is a simple Tavily-powered web search skill with no bundled executable code, but users should treat searches as external API requests.

Install only if you are comfortable sending search queries and selected search parameters to Tavily. Avoid using it for secrets, private internal content, regulated data, medical diagnosis, legal decisions, or tasks that do not actually require web search.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger condition is overly broad, stating the skill should be used whenever AI model invocation, intelligent dialogue, agent orchestration, or LLM applications are involved. In an agent ecosystem, this can cause the skill to activate outside intended search use cases and send unrelated user prompts to an external search API, increasing the chance of unintended data disclosure and inappropriate tool use.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill does not clearly warn that user queries will be transmitted to the third-party Tavily API. This is dangerous because users or upstream agents may pass sensitive prompts, internal project details, credentials, or regulated data into what appears to be a generic search workflow without informed consent about external disclosure.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.