v2.1.0

Mental Health Assistant

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 8:03 AM.

Analysis

This no-code mental-health support skill appears transparent and purpose-aligned, but users should treat it as supportive guidance only and be careful with crisis escalation and privacy.

GuidanceBefore installing, understand that this is a guidance and support skill, not a clinician. Do not rely on it for diagnosis, medication, or emergency care. If you or someone else is in immediate danger, contact local emergency services or a crisis hotline directly. If your agent has memory, messaging, or calling tools, approve any recording or outreach involving mental-health information carefully.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Human-Agent Trust Exploitation

SeverityLowConfidenceHighStatusNote

SKILL.md

专业心理健康支持助手，基于临床心理学与精神医学标准...非诊断工具，不能替代专业服务

The skill uses professional clinical framing in a sensitive mental-health context, while also including a clear disclaimer that it is not a diagnostic or professional-service substitute.

User impactUsers could over-trust the assistant because the topic is medical and crisis-related, even though the artifact does warn against using it as a replacement for professional help.

RecommendationUse this skill for education, emotional support, and self-help guidance only; rely on licensed professionals and emergency services for diagnosis, treatment, medication, or crises.

Tool Misuse and Exploitation

SeverityLowConfidenceHighStatusNote

references/crisis-resources.md

高风险 | 有具体计划、工具、时间地点 | 立即拨打热线/报警，不要独处

The skill instructs emergency escalation for high suicide risk; this is purpose-aligned, but would be a high-impact real-world action if used by an agent with calling or messaging tools.

User impactIf the user’s agent has communication capabilities, crisis escalation could involve contacting hotlines, emergency services, or other people.

RecommendationKeep any calls, messages, disclosures, or emergency escalation under clear user or human supervision, except where immediate safety procedures are legally and ethically required.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Memory and Context Poisoning

SeverityInfoConfidenceMediumStatusNote

references/ethics-guidelines.md

保密内容...对话内容...个人信息...测评结果...咨询记录

The reference material recognizes that mental-health conversations, personal information, assessment results, and records are sensitive; no actual storage mechanism is included.

User impactMental-health conversations may involve highly private information that should not be unnecessarily stored, summarized, or reused across unrelated tasks.

RecommendationAvoid sharing more personal detail than necessary, and only allow the agent to store or remember mental-health information when you explicitly want that.