ClawHub
Back to skill

Security audit

JSON检查工具

Security checks for vulnerabilities and agentic risk

Overview

This JSON linter is not clearly malicious, but it asks for broader agent powers than a simple JSON syntax checker needs.

Review this before installing. Use it only in a workspace where broad read/exec/write access is acceptable, and constrain it to JSON validation. Do not provide an API key or allow network/API behavior unless the publisher clarifies why it is required for linting.

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 (3)

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The skill is marketed as a simple JSON syntax validator, but its documentation also advertises file writing, external API integration, and command execution. This capability mismatch can mislead a user or orchestrator into granting broader permissions than necessary, increasing the attack surface and creating opportunities for unexpected side effects.

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The stated output contract describes a deterministic JSON scan report, while other sections describe system command execution and external API calls. This inconsistency obscures the real behavior of the skill, making it harder for users and policy systems to assess risk and potentially allowing risky actions under the guise of a harmless validator.

Vague Triggers

High
Confidence
95% confidence
Finding
The description says the skill should be used for broad efficiency, automation, batch processing, and workflow optimization tasks rather than narrowly for JSON linting. Overbroad routing conditions can cause an agent to invoke this skill in unrelated contexts, exposing read/exec/write capabilities where they are unnecessary and enabling misuse far beyond the advertised function.

Static analysis

No suspicious patterns detected.