ClawHub
Back to skill

Security audit

JSON格式化工具

Security checks for vulnerabilities and agentic risk

Overview

This JSON formatting skill is mostly a local utility, but it asks for and describes broader command, write, API, and credential-related capabilities that are not clearly needed for its stated purpose.

Review this skill carefully before installing. It appears intended for JSON formatting and validation, but it requests broader authority than that task needs. Avoid using it with secrets or private JSON unless you are comfortable with the agent having command execution, write access, and possible API/network handling; prefer a strictly local JSON formatter if you only need parsing or pretty-printing.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (4)

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The skill is presented as a JSON formatting/validation utility, but later advertises command execution capability unrelated to that purpose. Unnecessary exec capability expands the attack surface and can cause an agent to run shell commands based on untrusted JSON content or ambiguous user prompts, creating avoidable system-risk for a task that should be purely local parsing/formatting.

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
The documentation claims API integration, network behavior, and API key usage even though the core function is local JSON processing. This mismatch can mislead an agent into exfiltrating user-provided JSON to external services or requesting unnecessary credentials, increasing privacy and data-handling risk without a legitimate functional need.

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The skill's stated purpose is simple JSON formatting, validation, compression, and path extraction, but later sections broaden it to file I/O, API integration, and command execution. This inconsistency is dangerous because agents and users may trust the benign headline while hidden or conflicting sections enable higher-risk behaviors not reasonably expected from the advertised function.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill describes file handling, command execution, and external API behavior without prominent user-facing warnings about system changes, data exposure, or privacy impact. When risky capabilities are undocumented or underexplained, users may provide sensitive data or approve use under false assumptions, increasing the chance of unintended execution or exfiltration.

Static analysis

No suspicious patterns detected.