Security audit

知识管理工具（免费版）

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a knowledge-management prompt, but it requests command execution and describes broad, loosely scoped automation/API/file behaviors beyond its stated purpose.

Review this skill before installing. It does not show hidden code or clear malicious behavior, but it gives an agent broad command-execution and automation latitude for a loosely defined knowledge-management task. Use it only in a constrained workspace and confirm before allowing shell commands, package installs, external API calls, or file writes.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 96% confidence
Finding: The skill’s manifest and body describe a knowledge-management tool, but the invocation guidance and documented capabilities expand into unrelated design/visual-creation and generic automation use. This scope drift weakens user expectations and policy boundaries, increasing the chance the agent will invoke the skill in unintended contexts and perform actions outside its legitimate purpose.

Context-Inappropriate Capability

Medium

Confidence: 93% confidence
Finding: The skill declares exec and external API/network-adjacent behavior despite a stated purpose that mainly requires document analysis and template generation. Unnecessary command execution materially increases attack surface because a prompt-induced or confused invocation could lead to shell commands, network probing, or external data handling that the user would not expect from a knowledge-management skill.

Vague Triggers

High

Confidence: 97% confidence
Finding: The top-level description contains broad, conflicting activation guidance, including design/UI/poster/brand-visual scenarios unrelated to knowledge management. Overbroad routing instructions can cause the agent to invoke this skill for the wrong tasks, exposing users to command execution and file/system behaviors in contexts where they did not intend to grant such capabilities.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The example trigger phrase is underspecified and does not constrain what data sources, outputs, or side effects are allowed. Broad triggers increase the risk of accidental or prompt-manipulated activation, especially in a skill that advertises exec and external integration capability.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill describes file handling, API integration, and command execution, but it does not present a clear, prominent warning to users about side effects, data exposure, or system impact at the point of use. This creates a consent and transparency gap that can lead to unexpected writes, external data transmission, or local command execution.

Static analysis

No suspicious patterns detected.