Back to skill

Security audit

Mental Health Assistant

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese-language mental health support reference skill with crisis guidance, but it does not install code, persist data, access credentials, or hide unrelated behavior.

Install only if you want Chinese-language mental-health support and reference material. Do not treat it as therapy, diagnosis, medication advice, or emergency care; in imminent danger use local emergency services or a current crisis hotline, and verify hotline numbers for your region.

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
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (18)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill title and core description are written in Chinese and present the assistant as a general-purpose professional mental health support tool, but there is no indication that language is optional or that the skill is intentionally limited to Chinese-speaking users. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes broad everyday phrases such as '心情不好', '想聊聊', and '放松', which can activate this sensitive mental-health skill in contexts that are not actually seeking psychological support. In a crisis-oriented assistant, over-triggering can misroute ordinary conversations into mental-health workflows, causing inappropriate risk framing, unnecessary collection of sensitive emotional disclosures, or user distress.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill is defined entirely in Chinese without any language selection or consent mechanism, which can cause users to receive safety-critical mental-health guidance in a language they may not fully understand. In crisis and assessment scenarios, language mismatch increases the risk of misunderstanding suicide-risk questions, hotline guidance, or escalation instructions, potentially leading to harmful outcomes.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The document presents all instructions, warnings, and assessment guidance exclusively in Chinese, which can constitute a language/locale policy issue when no user opt-in or alternative language path is provided. The file does not state that the skill is intended only for a Chinese-speaking or region-specific audience.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This file provides structured mental-health techniques, including guidance relevant to depression, self-harm, and crisis-adjacent distress, but does not warn users that the material is informational and not a substitute for licensed care or emergency support. In a mental-health assistant context, users may over-rely on self-help content during acute deterioration, delaying crisis intervention or professional evaluation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file is entirely in Chinese and does not indicate any language selection or fallback behavior, which can cause users who do not read Chinese to misunderstand or miss crisis instructions, hotline numbers, and safety guidance. In a suicide and self-harm intervention context, loss of comprehension is safety-critical because delayed or incorrect action can materially increase risk during an emergency.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The file contains detailed DSM-5/ICD-11 diagnostic criteria, severity thresholds, differential diagnosis guidance, and tool-selection recommendations inside a mental-health support skill. Even with a disclaimer, this materially enables the agent to perform quasi-diagnostic triage beyond supportive assistance, which is risky in a high-stakes domain and could lead to misclassification, false reassurance, or inappropriate escalation decisions, especially around suicidality and severe depression.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The document title and all instructional content are presented only in Chinese, which constitutes a language/locale constraint in natural language. The file does not indicate that the skill is region-specific or provide any user opt-in or alternative language choice.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This guidance notes when to refer users with self-harm, suicide, psychosis, or severe impairment, but it does not provide a clear, immediate warning that the assistant is not sufficient for crisis situations or instruct users to seek urgent emergency or hotline support right away. In a mental-health skill, this omission is dangerous because users in acute crisis may continue a supportive conversation instead of escalating to time-sensitive professional help.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file provides breath-holding, deep-breathing, and muscle-tension exercises as broadly applicable calming tools without screening, contraindications, or stop conditions. In a mental-health skill, users may include people with panic disorder, asthma/COPD, cardiovascular issues, pregnancy, chronic pain, trauma sensitivity, or dizziness/fainting risk, so unsafeguarded instructions can worsen symptoms or cause physical distress.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This document provides detailed suicide risk assessment and intervention guidance, including crisis triage and scripts, but never clearly states that it is not a substitute for licensed clinical care or emergency services. In a mental-health assistant context, users or the agent may over-rely on the document during life-threatening crises, causing delayed escalation to emergency or professional support despite some hotline and emergency references being present.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

This is a real safety vulnerability: the document presents therapy modalities, explicitly includes populations with self-harm/suicide risk and mentions crisis intervention, but does not clearly state that the content is informational only and not a substitute for licensed clinical assessment or emergency support. In a mental-health assistant context, users may over-rely on these descriptions as actionable treatment guidance, delaying professional care or crisis escalation when acute risk is present.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The document is entirely written in Chinese and presents all user-facing guidance and example prompts in that language, with no indication that language choice is optional or limited to a justified region-specific context. Under the policy, a skill should not implicitly force a specific language or locale without user opt-in or clear documentation of the constraint.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

全文以中文编写并隐含要求按中文话术开展支持,但未说明这是面向中文用户的限定材料,也未给出语言切换或用户偏好选择。根据语言/地区政策要求,若技能默认强制单一语言而无用户选择,可能构成自然语言层面的策略不合规。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The document is entirely written in Chinese and does not indicate that language selection is optional or that alternative locales are available. Under the stated policy, natural-language content that enforces a specific language without user opt-in can be a locale-policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
98% confidence
Finding

The natural-language content forces a specific language/locale for all users, and there is no indication that the skill offers alternatives or that the Chinese-only constraint is intentional for a region-specific use case. This matches the policy category for language or locale restrictions without user opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill content is presented entirely in Chinese from the title onward, with no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking or region-specific audience. Under the policy rule, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The content forces a specific language/locale for all users through natural-language instructions and examples, but does not offer an opt-in choice or explain that the skill is intended only for Chinese-speaking or China-based contexts. Under the policy, language constraints should either be optional or clearly documented as region-specific and justified.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.