Security audit

JavaScript规范工具(免费版)

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to be a JavaScript style guide, but it includes broader tool and workflow authority than that purpose clearly requires.

Review this skill before installing. It does not show malicious behavior, but users should treat it as broader than a passive style guide: restrict it to JavaScript style review or snippet generation, avoid letting it run commands or modify files automatically, and be cautious with any deployment, network, or batch-processing use.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (6)

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The skill is presented as a JavaScript style guide, but these sections expand its scope into generic CRUD-style operations, runtime configuration, export/save/convert behavior, and workflow automation. This scope inflation can mislead an agent into treating a narrow advisory skill as an operational one, increasing the chance of unauthorized file, tool, or workflow actions beyond the user’s intent.

Description-Behavior Mismatch

Medium

Confidence: 90% confidence
Finding: The skill metadata says it is appropriate for development and deployment usage, while later sections say operations/deployment management is out of scope. Such contradictions create unsafe ambiguity for agent routing and may cause the skill to activate in higher-risk deployment contexts where a style-guide skill should not have authority or influence.

Intent-Code Divergence

Medium

Confidence: 95% confidence
Finding: The document claims the tool only provides suggestions and snippets, yet earlier sections advertise modify/delete/save/export style operations. This mismatch can cause an agent or user to underestimate the skill’s authority, leading to unexpected file changes or destructive actions under the guise of passive guidance.

Context-Inappropriate Capability

Low

Confidence: 80% confidence
Finding: The skill declares browser and broader network-oriented capabilities that are not justified by its stated purpose as a JavaScript style guide. Unnecessary capability signaling increases attack surface by encouraging tool use outside the minimum needed for code-style assistance.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The activation logic is overly broad, effectively matching many ordinary development conversations instead of specific JavaScript style-review requests. Over-triggering raises the risk that the skill will inject instructions, claim authority, or influence tool usage in unrelated tasks where its broader operational language becomes dangerous.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger phrase covers code generation, programming assistance, debugging, testing, and deployment, which is far wider than the skill’s stated advisory purpose. In context, this broad trigger interacts with contradictory capability claims and tool declarations, making unintended activation more likely in sensitive development workflows.

Static analysis

No suspicious patterns detected.