ClawHub

OpenClaw Enterprise

Security checks across malware telemetry and agentic risk

Overview

This skill appears purpose-aligned, with the main caution that broad triggers and routing could accidentally send business content into an external multi-agent workflow.

Install only if you want a broad enterprise workflow orchestrator. Use it deliberately, avoid sharing confidential business data unless your organization permits OpenAI or Anthropic processing, and prefer explicit invocation or confirmation before multi-agent routing.

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 (4)

Vague Triggers

Medium
Confidence
85% confidence
Finding
The trigger list contains very broad business phrases like '团队协作', '流程规划', and '企业自动化' that can match many normal conversations and cause unintended invocation. In a skill that may orchestrate multiple agents and has detected file access capabilities, accidental activation increases the chance of unnecessary data exposure to external APIs or unintended local actions.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The skill metadata and description use very broad activation language such as 'enterprise workflow automation' and 'multi-agent collaboration', which can match many ordinary business requests unrelated to this specific skill. Over-broad triggers increase the chance of accidental invocation, causing users to route sensitive enterprise tasks and data into a multi-agent workflow that may send content to external LLM APIs without deliberate intent.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger list includes highly generic phrases like 'AI employee', 'enterprise automation', 'chief of staff', and 'AI workflow', which overlap with common business language and are likely to be mentioned in benign conversations. In this skill's context, accidental activation is more concerning because it is positioned for enterprise operations and explicitly states that user task content may be sent to OpenAI or Anthropic, increasing data exposure risk from unintended routing.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The routing table uses broad, high-frequency business keywords such as '数据', '报告', '订单', and '客户' that can match many unrelated enterprise requests, making agent selection ambiguous. In a multi-agent enterprise orchestration context, this can misroute tasks, expose data to the wrong agent, and cause inappropriate actions or analysis based on the wrong business workflow.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal