Security audit

Json Lint Tool Free

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a local JSON linter, but its activation text is broader than its real purpose while it requests read, write, and command execution access.

Install only if you intend to use it for local JSON syntax validation. Before running it, specify a target directory and exclusions such as node_modules, .git, dist, secrets, and private data folders; do not rely on the broad analytics or visualization trigger text as an accurate description of its capabilities.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Behavioral ASTexec() Call, eval() Call, Dynamic Import
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

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 96% confidence
Finding: The manifest advertises broad data analysis, reporting, and visualization use cases that do not match a JSON syntax linter's documented behavior. This can cause the agent to invoke the skill in inappropriate contexts, leading to unnecessary workspace scanning and command execution under false pretenses.

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The capability coverage section expands the skill's scope to unsupported scenarios such as data analysis and visualization, which broadens activation beyond its real function. Overbroad capability claims are dangerous because they can route unrelated user requests into a skill that recursively scans files and may execute local commands.

Vague Triggers

High

Confidence: 97% confidence
Finding: Overly broad and mismatched trigger conditions materially increase the chance that the agent will auto-select this skill for unrelated requests. In this skill's context, that matters because it is authorized to read, write, and execute commands, so misrouting can expand file access and command execution beyond user intent.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The keyword-based activation scope is excessively broad and unconstrained, which can cause accidental invocation in many unrelated contexts. Because the skill performs recursive workspace scanning and may invoke command-line tooling, unnecessary activation increases exposure to unintended data access and execution.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill omits a clear warning that it may recursively scan the workspace and use exec-capable tooling. Without prominent disclosure, users and routing systems may underestimate the sensitivity of granting this skill access, increasing the risk of unintended broad file enumeration and local command execution.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.