ClawHub
Back to skill

Security audit

测试报告

Security checks across malware telemetry and agentic risk

Overview

This is a read-only QA reporting skill that formats test data into reports, with only minor risk that broad trigger phrases could activate it for unrelated report requests.

Before installing, be aware that this skill is intended for QA/test-reporting contexts. If you ask for a generic daily report or progress report, confirm that you want a QA-style report; review the separate larger QA skill set before running the advertised npx install command.

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
94% confidence
Finding
The trigger phrases include broad, common workplace expressions such as '日报', '周报', '报告模板', and '进度汇报', which can match many ordinary user requests outside the narrow intended QA-reporting context. This can cause unintended invocation of the skill, leading the agent to apply the wrong workflow, format, or assumptions to unrelated tasks and increasing the chance of misleading outputs or context confusion.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.