ClawHub
Back to skill

Security audit

JSON验证工具

Security checks for vulnerabilities and agentic risk

Overview

This JSON validation skill is not clearly malicious, but it asks for command execution, file writing, and API-backed use with broad and under-scoped instructions.

Review this skill before installing. Use it only for JSON validation tasks you explicitly choose, avoid giving it sensitive JSON or broad filesystem access, and confirm any command execution, file writes, or API calls before allowing them.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The skill claims that 'original risk code' was removed and external dependencies were cleaned up, yet it still declares powerful capabilities including exec, write, and API-backed use. This contradiction can mislead users and downstream agents into underestimating the real attack surface, increasing the chance that command execution, file modification, or networked actions are approved without appropriate scrutiny.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The description uses broad invocation language such as using the skill whenever API integration, webhooks, system connections, or workflow scenarios are involved, without defining strict boundaries or safe trigger conditions. In agent ecosystems, vague applicability increases the likelihood of over-invocation in contexts where untrusted input may be passed into privileged operations like exec or write.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill advertises file handling, API integration, and command execution capabilities, but does not present a prominent up-front warning about their operational and security impact before users reach those sections. Because these are high-risk primitives, understated disclosure can cause users or agents to treat the skill as a simple validator while it actually has the ability to alter files, run commands, and communicate externally.

Static analysis

No suspicious patterns detected.