ClawHub

Mind Companion

Security checks across malware telemetry and agentic risk

Overview

This appears to be an emotional-support skill, but the available evidence shows unclear safety boundaries for crisis or self-harm situations.

Review this skill carefully before installing. It may be fine for casual companionship, but it should not be treated as therapy, crisis support, or emergency help unless the publisher adds clear boundaries and escalation instructions for self-harm, abuse, or imminent danger.

SkillSpector

By NVIDIA
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
92% confidence
Finding
The skill's activation cues are broad phrases like 'someone to talk to,' 'feeling down,' and 'just want to talk,' which are common in ordinary conversation and can cause the skill to trigger in situations the user did not explicitly intend. In a mental-health-adjacent skill, accidental invocation is more concerning because it may steer vulnerable users into emotionally sensitive interactions without clear consent or boundaries.

Missing User Warnings

High
Confidence
97% confidence
Finding
This skill presents itself as a 'safe, warm, judgment-free listener' for users who have troubles they 'can't share with anyone else,' but it provides no crisis limitations, safety warnings, or escalation instructions for self-harm, suicidality, abuse, or other acute mental-health risks. That omission is dangerous because vulnerable users may rely on the skill as a substitute for urgent human or professional help, delaying intervention in high-risk situations.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal