Back to skill

Security audit

市场机会侦察兵

Security checks for vulnerabilities and agentic risk

Overview

The skill presents itself as live financial monitoring, but its reports are mostly hard-coded current-dated market claims, so users could mistake sample data for real investment signals.

Review this carefully before installing. It does not appear to steal data or install a backdoor, but it should not be treated as a reliable market-monitoring tool until live data parsing, source timestamps, failure handling, and dependency pinning are added. Avoid using its output for investment decisions in its current form.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (2)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:117
Finding

Unpinned Third-Party Dependencies Permit Mutable Supply-Chain Code

Content
View full analysis
Remediation
View remediation

other

Error
Location
scout.py:21
Finding

Current-Dated Financial Report Contains Hard-Coded and Unvalidated Market Claims

Content
View full analysis
/dev/null") # 恒生指数 hs = run_command("curl -s 'https://push2.eastmoney.com/api/qt/stock/get?secid=100.HSI&fields=f43,f44,f170' 2>/dev/null") print(f"- 上证指数:数据获取中...") print(f"- 恒生指数:数据获取中...") print() ``` The substantive financial content is instead supplied through fixed literals: ```python def get_hot_sectors(): """获取热门板块""" print("## 🔥 热门板块\n") sectors = [ ("白酒", "+2.5%"), ("新能源", "+1.8%"), ("人工智能", "+3.2%"), ("半导体", "+1.5%"), ("医药", "-0.5%") ] for i, (name, change) in enumerate(sectors, 1): print(f"{i}. {name} ({change})") print() def get_unusual_stocks(): """获取异动个股""" print("## ⚡ 异动个股\n") stocks = [ ("贵州茅台", "600519", "+5.2%", "放量突破"), ("腾讯控股", "00700", "+3.8%", "北向资金流入"), ("阿里巴巴", "09988", "+2.1%", "超跌反弹"), ("宁德时代", "300750", "+4.5%", "突破新高") ] for name, code, change, reason in stocks: print(f"- {name}({code}):{change} ({reason})") print() def get_finance_news(): """获取财经新闻""" print("## 📰 重要新闻\n") news = [ "央行:保持流动性合理充裕 - 财联社", "多地出台房地产新政 - 证券时报", "AI 大模型竞争加剧 - 36 氪", "新能源汽车销量创新高 - 界面新闻" ] for i, item in enumerate(news, 1): print(f"{i}. {item}") print() def get_opportunity_hints(): """生成机会提示""" print("## 💡 机会提示\n") hints = [ "- 白酒板块连续 3 日资金净流入", "- 人工智能板块突破 60 日均线", "- 北向资金今日净流入超 50 亿", ...[truncated 2452 chars]
Remediation
View remediation
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (8)

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill defines a generic helper that can execute arbitrary shell commands despite its advertised purpose being market monitoring and report generation. This capability is over-privileged for the task and creates a dangerous primitive that could be repurposed for command execution if the code is modified, extended, or integrated with external inputs.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
97% confidence
Finding

This finding is valid because the skill exposes a tool-like primitive that executes shell commands with shell=True, a classic avenue for tool-parameter abuse and command injection. Even though current calls are hardcoded curl commands, the existence of this primitive in a market-reporting skill is unjustified and becomes more dangerous if any argument source becomes user- or data-controlled.

Content

Scanner excerpt · scout.py (reported line 14)May include surrounding context.

python
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}"

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding

The skill advertises shell-based installation and runtime commands, plus cron execution and pip dependency installation, but does not declare any explicit tool scope such as permissions or allowed-tools. This creates an ambiguity gap where an agent or platform may permit shell execution more broadly than users expect, increasing the risk of unintended command execution or unsafe automation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This code file contains natural-language descriptions and CLI help that force a specific language/locale for all users. Under the policy, a skill should not mandate a language without user opt-in unless the locale constraint is clearly documented and justified.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
96% confidence
Finding

The code invokes subprocess.run with shell=True, which is inherently risky because any future or indirect use of untrusted input in cmd can lead to shell injection and arbitrary command execution. In this skill’s context, shell execution is unnecessary for ordinary market-data retrieval and expands the attack surface beyond the stated functionality.

Content

Scanner excerpt · scout.py (reported line 14)May include surrounding context.

python
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}"

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest says the skill automatically monitors A/H-share anomalies, financial news, and hot sectors to generate investment opportunity briefs. Aside from two curl calls whose results are not parsed or used, the report is built from fixed hard-coded sectors, stocks, news, and hints, so the actual behavior does not match the claimed monitoring capability.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

Several docstrings say the code 'gets' hot sectors, unusual stocks, finance news, and opportunity hints, and the module docstring says it 'automatically monitors' market movements. In reality, these functions print static, hard-coded lists rather than retrieving or monitoring real-time data, which contradicts the documented intent rather than merely omitting implementation detail.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The manifest description and the user-facing README content consistently force Chinese-language interaction and documentation. Under the policy, language constraints should either be optional for the user or clearly justified as region-specific; this file does neither explicitly.

Content

No source excerpt is available for this finding.