ClawHub
Back to skill

Security audit

Expense Categorization Optimizer

Security checks across malware telemetry and agentic risk

Overview

This skill is a descriptive expense-categorization helper with no evidence of account access, transactions, network use, persistence, or hidden behavior.

Reasonable to install if you want template-style expense categorization help. Treat its financial, tax, pricing, and compliance suggestions as general information only, and verify important decisions with a qualified professional.

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 (3)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The example prompts expand the skill far beyond expense categorization into broad financial analysis, tax optimization, invoicing, pricing, and report interpretation. In a routing or auto-invocation system, this can cause the skill to be selected for unrelated financial tasks, producing out-of-scope guidance and increasing the chance of unsafe or misleading advice.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger language is generic and overlaps heavily with other financial-planning skills, making invocation boundaries unclear. When a skill with advisory content activates too broadly, users may receive inappropriate templates or recommendations for tasks requiring different controls, expertise, or disclaimers.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger keywords are excessively generic, including common words like 'expense', 'categorization', and 'optimizer'. This can cause the skill to activate unintentionally during ordinary user conversations, creating prompt-routing confusion and increasing the chance that users invoke the skill without clear intent.

VirusTotal

58/58 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.