ClawHub
Back to skill

Security audit

Mental Health Psychoeducation

Security checks for vulnerabilities and agentic risk

Overview

This is a static educational mental-health skill with clear professional-care disclaimers and no code, persistence, credential use, or hidden execution behavior.

Install only with the understanding that this is educational mental-health information, not medical advice, diagnosis, therapy, medication guidance, or crisis support. In crisis situations, users should contact local emergency services or an appropriate local crisis line; US users can call or text 988.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Anti-Refusal Statement

High
Category
Anti-Refusal
Content
Developed for borderline personality disorder, now used for emotion regulation struggles. Balances acceptance and change — you validate your feelings while also learning skills to manage them.

**Four skill modules:**
1. **Mindfulness:** Stay present, observe without judgment
2. **Distress Tolerance:** Survive crises without making things worse (self-harm, substance use, impulsive actions)
3. **Emotion Regulation:** Understand and manage intense emotions
4. **Interpersonal Effectiveness:** Communicate needs, set boundaries, maintain relationships
Confidence
80% confidence
Finding
Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger list includes very broad, high-frequency phrases such as 'mental health', 'anxiety', and 'depression', which can cause the skill to activate in many ordinary conversations beyond explicit psychoeducation requests. In a sensitive domain, unintended invocation can steer users into quasi-clinical content when they did not ask for it, increasing the chance of overreach or inappropriate guidance.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The crisis guidance prominently instructs users to call 988 or local emergency services, but does not first determine the user's country or region. For users outside the US, US-centric crisis instructions may be confusing or unusable during an emergency, which is especially risky given the mental health and suicidality context.

Static analysis

No suspicious patterns detected.