Security audit

麦肯锡思考框架全集

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only consulting-framework reference skill with no code execution, credential access, persistence, or hidden data handling.

Safe to install if you want a prompt-only business-framework reference. Use it deliberately for consulting-style analysis, specify your preferred language when needed, and avoid entering confidential business data unless your normal agent environment is trusted.

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The trigger examples are broad enough that ordinary user requests like '帮我分析这个市场' could activate the skill without clear user intent to invoke this specific framework pack. That can cause unintended behavior selection, override user preference for another method, or create prompt-routing ambiguity across skills.

Natural-Language Policy Violations

Medium

Confidence: 78% confidence
Finding: The skill is written entirely in Chinese and presents its usage patterns and outputs as Chinese-first without indicating language adaptation based on user preference. While not a direct security exploit, this can lead to unauthorized language switching, reduced user comprehension, and prompt-control issues when the surrounding system expects the assistant to preserve the user's language.

Static analysis

No suspicious patterns detected.