Security audit

AI Decision Coach - 决策教练

Security checks for vulnerabilities and agentic risk

Overview

This is a single Markdown decision-analysis skill with no code execution, network access, file access, or persistence; its main caveat is broad activation and Chinese-only presentation.

Install only if you want a Chinese decision-analysis template. For medical, legal, financial, crisis, or other high-stakes choices, treat its output as a general reflection aid and use qualified specialist advice rather than relying on this skill alone.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

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)

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: The skill metadata and content are written in Chinese and do not indicate language negotiation or fallback based on the user's preference. Forcing a specific language can degrade user understanding of recommendations, which is especially problematic in decision-support contexts where nuance and risk communication matter.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The trigger condition '当用户面临选择困境、纠结、犹豫不决时' is broad and subjective, so the skill may activate for many loosely related conversations. Over-broad activation can cause unintended routing, leading the assistant to apply this framework when a more specialized or safer skill should handle the request.

Static analysis

No suspicious patterns detected.