ClawHub
Back to skill

Security audit

财务分析入门工具

Security checks for vulnerabilities and agentic risk

Overview

This is a finance-analysis Markdown skill with disclosed local read/write/exec capabilities and no artifact evidence of hidden, destructive, or exfiltrating behavior.

Before installing, treat this as a local finance-analysis aid and use it only with financial data you intend the agent to read or process. Be cautious with optional external financial data APIs and any command execution; provide credentials through environment variables only when needed and approve generated scripts or file writes deliberately.

SkillSpector

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
92% confidence
Finding
The invocation text says to use the skill whenever the user needs accounting/finance-tool-related functionality, which is overly broad and may cause the agent to activate this skill for vague or only loosely related requests. Over-broad triggers can lead to unintended tool use, including unnecessary read/write/exec access in contexts where a safer or narrower skill should have been selected.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The capability-scope section includes very broad trigger phrases like natural-language references to data analysis, report generation, statistical insight, and visualization, which extend far beyond the stated finance-focused purpose. In a skill with read/exec/write tools, this can cause accidental routing of generic analytical requests into a more privileged skill, increasing the chance of unnecessary command execution or data handling.

Static analysis

No suspicious patterns detected.