ClawHub
Back to skill

Security audit

JSON解析器

Security checks for vulnerabilities and agentic risk

Overview

This JSON parser skill mainly describes normal data conversion, but it asks for command execution and an unspecified API key without clear limits.

Review before installing. Use this only in a constrained workspace, avoid giving it secrets or broad project access unless needed, and require confirmation before any shell command or file write. The skill would be safer if it documented exact allowed commands and why an API key is needed.

SkillSpector

Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The skill is presented as a JSON parsing/validation utility, yet its documentation acknowledges command execution capability. That mismatch can mislead users and host agents into granting broader permissions than necessary, increasing the chance that untrusted input is routed into shell execution or other unsafe actions.

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The skill claims to be a JSON parser, but the documented runtime model depends on an LLM, API key handling, and executable behavior unrelated to simple parsing. This overbroad capability surface violates least privilege and can cause agents or users to trust the skill with secrets and execution rights they would not grant to a normal parser.

Static analysis

No suspicious patterns detected.