ClawHub
Back to skill

Security audit

Json Toolkit Free

Security checks across malware telemetry and agentic risk

Overview

This skill is mostly a JSON best-practices document, but it requests write and command execution authority that its own instructions say it does not need.

Review this skill before installing. It appears to be a JSON guidance document, but it asks for file-writing and command-execution permissions that are not necessary for that purpose. Prefer a version that is read-only, or install only if you are comfortable with those broader permissions and can enforce confirmation for any file changes or shell commands.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The skill is presented as a defensive JSON best-practices guide, but these sections describe operational behaviors such as create/query/export, modify/reset/import, and save/convert. That mismatch can cause an agent or user to over-trust the skill and allow broader data-handling actions than expected, increasing the chance of unintended file or workflow manipulation.

Intent-Code Divergence

High
Confidence
97% confidence
Finding
The document claims the skill is pure Markdown and does not require exec, yet the manifest exposes exec capability. This creates a dangerous trust gap: operators may approve the skill as passive documentation while the agent runtime still grants command-execution privileges that could be abused if invoked through ambiguous instructions or future edits.

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The skill declares write and exec even though its stated purpose is a Markdown guide for JSON conventions. Excess capabilities violate least privilege and expand the attack surface, enabling file modification or command execution paths unrelated to the advertised function.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill exposes write and exec capabilities without warning users that system-changing actions may occur. In the context of a supposedly instructional JSON skill, that omission makes the capability set more dangerous because users are less likely to expect or scrutinize potentially destructive operations.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.