ClawHub
Back to skill

Security audit

Security Intelligent Cs

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only banking customer-service skill, but real deployments should add stricter privacy and identity-verification controls.

Before using this with real customers, require identity verification through official bank channels, collect only the minimum needed information, never request full account numbers or passwords, mask sensitive fields, and route complaints into an approved secure ticketing system.

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
89% confidence
Finding
The trigger list is very broad and consists of common banking and support phrases, which can cause the skill to activate in contexts where the user did not intend to invoke this workflow. In a financial-service setting, accidental activation is more dangerous because the skill may steer conversations into account, transaction, or complaint flows that solicit sensitive information without sufficient scoping or authorization checks.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The complaint-handling SOP instructs collection of personal and account-related data, including name, account/card last four digits, event timing, and amount, but does not include any privacy notice, minimization rule, masking standard, retention limit, or authentication requirement. In a banking context this materially increases the risk of oversharing, unsafe logging, social-engineering abuse, and noncompliant handling of regulated financial data.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.