ClawHub
Back to skill

Security audit

市场机会侦察兵

Security checks for vulnerabilities and agentic risk

Overview

This skill is a market-report generator that makes limited public market-data requests and does not show credential access, destructive behavior, or hidden persistence.

Install only if you are comfortable with a local script making public market-data network requests. Avoid adding untrusted input to the shell helper, review or pin any manually installed Python dependencies, and only add the cron entry if you want recurring reports.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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 (2)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
def run_command(cmd):
    """执行 shell 命令并返回结果"""
    try:
        result = subprocess.run(cmd, shell=True, capture_output=True, text=True, timeout=30)
        return result.stdout.strip()
    except Exception as e:
        return f"Error: {e}"
Confidence
97% confidence
Finding
The code executes shell commands through subprocess.run with shell=True, which is inherently dangerous because it invokes a shell interpreter and expands metacharacters, redirects, and other shell syntax. In this file the current call sites use hardcoded curl commands, but the helper is generic and enables arbitrary command execution if reused with user-controlled input later, which is unjustified for a market-reporting skill.

Context-Inappropriate Capability

High
Confidence
98% confidence
Finding
The helper function run_command is a generic arbitrary shell execution primitive, which is broader than needed for a skill that only gathers market data and prints a report. This creates unnecessary attack surface: any future modification or indirect input flow into cmd could turn the skill into a command-execution vehicle.

Static analysis

No suspicious patterns detected.