ClawHub
Back to skill

Security audit

Finance Acct Tool Free

Security checks across malware telemetry and agentic risk

Overview

This finance skill is mostly accounting-oriented, but it under-discloses sensitive data flow and execution scope risks that users should review before installing.

Install only if you are comfortable granting read/write/exec access for local financial workflows. Do not provide a callback_url unless you understand exactly what data may be sent externally, and verify what finance.py will execute before letting the agent run the documented commands.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
89% confidence
Finding
The skill repeatedly claims that all financial data remains local and that no external API calls are involved, yet it documents a callback_url parameter for asynchronous notifications. In a finance tool, that inconsistency can cause users or agents to send accounting results or metadata to an external endpoint under the false assumption that the workflow is strictly local.

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
This is a true security-relevant documentation mismatch: the skill says financial data stays local, but the callback_url field implies external delivery of processing results. Because the skill handles sensitive financial records, even optional outbound callbacks can create confidentiality and compliance risks if users are not clearly warned.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The activation language is overly broad, including generic data analysis, reporting, statistical insight, and visualization scenarios beyond narrow accounting tasks. That can cause the skill to be invoked for unrelated requests, increasing the chance that sensitive user data is routed into a finance-oriented exec-enabled workflow unnecessarily.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The capability coverage section lists a long set of broad scene keywords with weak boundaries, effectively advertising the skill as applicable to many loosely related tasks. In an exec-capable skill, ambiguous routing increases the risk of over-collection, accidental execution, or misuse on data outside the intended financial accounting context.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.