Security audit

都市治愈空间

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only calming companion skill with some wellness-safety caveats to consider, but no hidden execution, data access, persistence, or destructive behavior.

Installers should treat this as a gentle Chinese-language relaxation and companionship skill, not professional mental-health care. Users with acute distress, self-harm concerns, panic, dizziness, respiratory sensitivity, or fire-safety constraints should rely on appropriate professional or emergency support and use any breathing, candle, incense, or aromatherapy suggestions cautiously.

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 (5)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The skill’s trigger conditions are extremely broad and map to common emotional states such as fatigue, stress, wanting to relax, or late-night conversation. That makes the skill likely to intercept many ordinary user requests and route them into a comfort-focused persona, which can crowd out more appropriate skills and, in some cases, mishandle users with significant emotional distress by keeping them in a non-specialist support flow.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This markdown file is entirely written in Chinese and provides no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking audience. Under the policy rule for language/locale, a skill that implicitly forces one language without opt-in can be a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The natural-language instructions, examples, and persona are entirely in Chinese, which implicitly constrains the interaction language. Because no user opt-in or language-choice mechanism is provided, this may violate a language/locale policy requiring user choice unless the locale restriction is explicitly documented and justified.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The file provides breathing and meditation instructions to emotionally vulnerable users without warning them to stop if they feel dizzy, distressed, or uncomfortable, and without clarifying that these practices are not a substitute for professional medical or mental health care. In a soothing-support skill, users may over-rely on the guidance during panic, trauma, or other health-sensitive situations, increasing risk of harm or delayed escalation to appropriate care.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The recommendation to light a candle or use incense/aromatherapy lacks basic precautions about fire safety, smoke sensitivity, and ventilation. In a late-night calming context, users may be tired or emotionally distracted, which makes unattended flame or poorly ventilated incense use more hazardous.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.