ClawHub
Back to skill

Security audit

Agent群组工具专业版

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate enterprise group-management skill, but it asks for broad invocation plus command, write, bot, and external-sync authority that needs human review before use.

Install only if you intend to use it for enterprise multi-agent group governance. Before enabling it, narrow when the skill can trigger, require confirmation for command execution, permission changes, exports, Webhooks, IM sync, and SIEM/BI forwarding, and verify destination allowlists, audit logging, retention, and credential handling.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

High
Confidence
94% confidence
Finding
The trigger condition '需要AI模型调用、智能对话、Agent编排、LLM应用时使用' is extremely broad and can match many unrelated workflows, increasing the chance the skill is invoked in contexts involving sensitive data or system actions. Because this skill also advertises exec, write, external integrations, and data export capabilities, unintended invocation could cause unauthorized command execution, outbound transmission, or modification of local resources.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill describes capabilities for Webhook/API integration, SIEM/BI export, bot-driven external calls, analytics generation, and command execution, but the warnings are not prominent relative to the powerful functionality. In practice, a user or orchestrator could activate features that move sensitive group messages, audit data, or system outputs to external endpoints without sufficiently explicit informed consent or impact disclosure.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.