ClawHub

@kanyun/rush-find-skills

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed helper for finding and installing other agent skills, with normal package-manager risk but no evidence of hidden or malicious behavior.

Install this only if you want your agent to search a reskill registry and help install other skills. Before approving any install, check the registry URL, skill name, publisher, version, and target agent, and prefer a trusted local reskill installation over automatically running the latest npx package.

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
91% confidence
Finding
The top-level description activates on very broad user phrases like 'how do I do X' and 'can ...', which are common in ordinary help requests unrelated to skill discovery. That can cause the skill to trigger too often and steer the agent toward registry search and potential package installation in situations where the user only wanted direct assistance, increasing exposure to unneeded external package recommendations and follow-on command execution.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The 'When to Use This Skill' section includes ambiguous triggers such as general 'how do I do X', 'can you do X', and expressions of interest in extending capabilities, without firm boundaries. In this skill's context, those broad conditions are more dangerous because the workflow leads from discovery into CLI-based registry interaction and possible installation, so an overbroad trigger can escalate ordinary conversation into external package operations.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal