Security audit

jiusansqmy

Security checks for vulnerabilities and agentic risk

Overview

This is a document-writing guide with no code, network access, credential use, persistence, or hidden high-impact behavior.

Installers should treat this as a specialized Chinese public-policy document drafting aid. Be aware that it may activate on broad topic mentions and default to a formal official style, so use explicit instructions if you want a different tone, format, or only general discussion.

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: 95% confidence
Finding: The trigger description is broad enough to activate on common mentions of topics like '九三学社', '社情民意', or '参政议政', which can cause the skill to engage outside a user's actual intent to use this specialized workflow. This increases the chance of unintended steering of responses toward a rigid political-writing format, reducing user control and potentially causing inappropriate handling of adjacent requests.

Natural-Language Policy Violations

Medium

Confidence: 88% confidence
Finding: The skill imposes a specific rhetorical style and official-sounding writing mode as a requirement, without indicating that the user can opt out or request a different tone. While not directly a security exploit, this can override user intent and lead to unauthorized style steering, especially if the skill is activated unintentionally due to the broad trigger logic.

Static analysis

No suspicious patterns detected.