ClawHub
Back to skill

Security audit

Clawmatch

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only compatibility-matching skill with no code or permissions, though it may prompt users to share sensitive personal profiling details.

This looks safe to install as a coming-soon, documentation-only skill. Before using it, remember that useful compatibility advice may require sensitive personal details; share only information you are comfortable putting into your agent session, and avoid entering private details about other people without their consent.

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 (1)

Vague Triggers

Medium
Confidence
82% confidence
Finding
The skill advertises broad trigger phrases like "find my match" and "who am I compatible with" without defining scope boundaries, consent requirements, or what data sources may be used. In an agent ecosystem, this can cause over-broad invocation on sensitive relationship, hiring, or profiling requests and may lead the agent to engage in high-risk psychological inference without clear user intent or safeguards.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.