ClawHub
Back to skill

Security audit

Accounting Finance Tool Free

Security checks across malware telemetry and agentic risk

Overview

The skill is a finance-analysis Markdown skill, but it asks for broad read/write/exec authority and includes under-scoped callback and network-related behavior that users should review before installing.

Review this skill before installing. It appears to be a finance-analysis helper rather than malware, but install it only if you are comfortable with an agent using read/write/exec for finance tasks. Avoid supplying callback URLs or external data-source credentials unless you understand what will be sent and explicitly approve any command or network action.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill defines a `callback_url` input and references network testing despite being positioned as a local financial-analysis tool with no stated need for outbound callbacks. This expands the trust boundary unnecessarily and could enable unreviewed data exfiltration or SSRF-like behavior if an agent posts results or probes arbitrary URLs supplied in user input.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The activation guidance uses broad triggers such as '需要数据分析、报表生成、统计洞察、数据可视化时使用,' which are far wider than the actual finance-specific scope. Overbroad routing can cause the skill to activate for unrelated tasks and inherit access to `read`, `write`, and `exec`, increasing the chance of unnecessary privileged actions on loosely related prompts.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The example trigger '帮我分析一下贵州茅台的财务状况' is generic enough that an agent may route many broad company-analysis prompts into an exec-enabled skill without clarifying data sources, boundaries, or side effects. In context, this matters because the skill advertises natural-language execution flows and privileged tools, so ambiguous activation can lead to overreach.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The documentation explicitly states the skill is 'MD+EXEC' and may use command execution, but it does not warn users about execution side effects, filesystem changes, subprocess risks, or possible network interactions. In an agent environment, silent exec capability materially raises the risk of unintended command execution from natural-language prompts or derived workflow steps.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.