Equity Analyst

• Outcome: Pass
• Latest audit: Jul 10, 2026 · 4:23 AM UTC
• Version: 1.0.7

Overview

This is a Korean stock-analysis skill with higher-risk helper scripts, but the inspected behavior stays aligned with scraping public Naver Finance data and producing local reports.

Install only if you are comfortable with optional Python scripts that scrape Naver Finance, control an OpenClaw browser profile, run local analysis code, and save report files. Review the scripts before running batch reports, use a clean browser profile if possible, and do not enable any scheduled or LINE delivery workflow without reviewing and configuring it yourself.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Output Handling: Unvalidated Output Injection, Cross-Context Output, Unbounded Output
• Tool Misuse: Tool Parameter Abuse, Chaining Abuse, Unsafe Defaults
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration

Findings (7)

subprocess module call

Medium

• Category: Dangerous Code Execution

Content

def run_cmd(cmd, timeout=30):
    """Run shell command and return output."""
    result = subprocess.run(
        cmd,
        shell=True,
        capture_output=True,

• Confidence: 98% confidence
• Finding: result = subprocess.run( cmd, shell=True, capture_output=True, text=True, timeout=timeout )

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 91% confidence
• Finding: The skill advertises no declared permissions, yet the content indicates code/scripts with file read/write, network access, and shell-capable automation. That mismatch is dangerous because it removes transparency and prevents appropriate sandboxing or informed approval for actions that can exfiltrate data, modify local files, or run unintended commands.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence
• Finding: The documented behavior describes a narrow single-stock analysis skill, but the referenced implementation scope is broader: batch reporting, direct scraping/automation, and local file output. This discrepancy is security-relevant because users and reviewers may approve a seemingly limited analytical skill while hidden or undocumented behaviors perform larger-scale data collection, persistence, or automation.

Context-Inappropriate Capability

Medium

• Confidence: 90% confidence
• Finding: Driving an external browser through shell-command automation gives the skill host-level process execution capability beyond a normal equity-analysis role. In this file, that capability is coupled with shell command construction and remote web content, which increases the chance of abuse or unintended local command execution.

Context-Inappropriate Capability

High

• Confidence: 88% confidence
• Finding: Executing a separate local Python program extends the skill from data analysis into arbitrary local code execution, which is outside the declared role and broadens the trusted computing base. If analyze.py or the Python interpreter path is modified, hijacked, or replaced, the script will execute attacker-controlled code on the host.

Unvalidated Output Injection

High

• Category: Output Handling

Content

def run_cmd(cmd, timeout=30):
    """Run shell command and return output."""
    result = subprocess.run(
        cmd,
        shell=True,
        capture_output=True,

• Confidence: 97% confidence
• Finding: subprocess.run( cmd, shell=True, capture_output

Tool Parameter Abuse

High

• Category: Tool Misuse

Content

def run_cmd(cmd, timeout=30):
    """Run shell command and return output."""
    result = subprocess.run(
        cmd,
        shell=True,
        capture_output=True,

• Confidence: 98% confidence
• Finding: subprocess.run( cmd, shell=True

VirusTotal

66/66 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.