ClawHub
Back to skill

Security audit

Finance Customer Service Hub

Security checks across malware telemetry and agentic risk

Overview

This is a text-only financial customer-service framework skill with no executable code, credential access, or hidden behavior, though its broad triggers need careful configuration.

Safe to install for drafting customer-service knowledge-base structures, routing flows, and general support frameworks. Configure narrow activation rules where possible, and do not rely on it for regulated financial advice, account-specific support, or authenticated customer actions without approved systems and human review.

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
94% confidence
Finding
The trigger phrases are very broad, generic terms such as 'knowledge management', 'fintech', and 'financial customer service' that are likely to match ordinary financial-support requests outside the intended scope. In an agent-routing context, this can cause the skill to activate unintentionally, intercept user traffic, and answer regulated financial-service questions inappropriately or without proper domain-specific safeguards.

Natural-Language Policy Violations

Medium
Confidence
86% confidence
Finding
The phrase '中文触发词（优先）' indicates the skill gives priority to Chinese triggers without explicit user language preference or consent. This can cause unintended routing or response-language mismatches, especially in multilingual environments, which increases the chance of confusing users and mishandling financial-support interactions.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.