ClawHub
Back to skill

Security audit

DNS查询工具

Security checks for vulnerabilities and agentic risk

Overview

This DNS lookup skill is not malicious, but it asks for broad command and file authority without clearly limiting how those powers are used.

Review before installing. Use it only in an agent environment where command execution is controlled, and provide simple DNS lookup requests rather than arbitrary instructions. The publisher should narrow the manifest to DNS-specific parameters, document exact command behavior, and remove unnecessary write access.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The skill is presented as a DNS lookup utility, but its documented interface accepts unconstrained free-form instruction and context fields instead of narrowly scoped DNS parameters such as hostname, record type, or server. In a skill that has exec capability, this ambiguity increases the chance that downstream implementations interpret arbitrary user text as executable command input, expanding the attack surface beyond simple DNS queries.

Intent-Code Divergence

Medium
Confidence
86% confidence
Finding
The documented response format does not describe DNS resolution results and instead returns a vague configuration-oriented result field. This mismatch can hide what the skill actually does, impair validation and monitoring, and make it easier for unsafe or unexpected behavior to pass unnoticed in an agent workflow.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The description uses broad, ambiguous invocation language unrelated to the narrow DNS-lookup purpose, including generic automation and marketing-style applicability claims. Overbroad positioning can cause agents or users to invoke the skill in unintended contexts, and when combined with read/exec/write permissions, this raises the risk of misuse or unsafe prompt-to-command behavior.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill advertises an execute-capable model (MD+execute()) without an explicit warning that it may run system commands or clarifying the limits of that execution. In agent environments, undocumented command execution materially increases risk because users may provide untrusted input expecting a harmless lookup while the implementation could invoke shell commands behind the scenes.

Static analysis

No suspicious patterns detected.