Back to skill

Security audit

Stock Analysis Framework

Security checks for vulnerabilities and agentic risk

Overview

This is a stock-analysis helper whose network calls and portfolio reading fit its stated purpose, with privacy and investment-advice caveats users should understand.

Install only in an environment where you are comfortable exposing queried ticker symbols to third-party market-data services. Treat all outputs as informational, not investment advice, and review any local portfolio file before running the monitoring or briefing scripts.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (16)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

该描述明显高于代码实际能力。代码的主要用途是对单只A股做一次性的命令行行情/技术/资金概览,属于简化版分析脚本,而不是“完整的六维分析框架及盯盘监测工具”。六维中只有技术面、资金面和部分心理面有实际计算;消息面和宏观面基本是提示性质,基本面完全缺失。描述中提到的“实时盯盘脚本、简报生成工具、持仓监控、交易策略制定”在该代码片段中均未体现。因此存在实质性的描述与行为不一致。

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

声明描述的是功能全面的A股分析/监控框架,但实际代码只是一个单文件简报工具,核心行为是读取本地投资组合、请求实时行情并打印持仓表现摘要。它没有执行基本面、技术面、资金面、情绪面或宏观政策分析,也没有实现盯盘监测循环、告警、模板化分析输出或策略分析。虽然“简报生成工具”和“持仓监控”与声明有部分重合,但整体主用途和能力范围明显小于且不同于声明,因此属于显著描述-行为不匹配。

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

声明描述的是一个功能完整的股票分析与监控框架,但提供的代码片段只包含静态股票名称-代码映射和几个简单查询/格式化函数。该代码没有进行任何基本面、技术面、消息面、资金面、情绪面或宏观政策分析,也没有监控、抓取、计算、报警、生成报告等行为。虽然股票代码映射可以作为此类系统的底层辅助组件,但就该代码片段本身而言,其实际功能与声明的核心用途存在明显且实质性的差距,因此应判定为描述与行为不匹配。

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

该代码与声明中的“实时盯盘监测工具”部分基本一致,但与“完整的A股股票分析框架(六维体系)”“结构化分析模板”“简报生成工具”等核心描述明显不符。代码没有实现六大分析维度中的绝大多数内容,也没有深度分析、策略制定或报告生成功能。其主要功能只是一个行情监控与异动告警脚本,因此声明显著夸大了实际能力,应判定为描述与行为不匹配。

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding

The skill advertises or implies network-capable scripts but does not declare any explicit tool scope such as permissions or allowed-tools. This creates an avoidable trust and containment gap: a host or user may not realize the skill can reach external market data sources, increasing the risk of unintended data egress or over-privileged execution.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill explicitly supports portfolio monitoring and briefing generation, which can involve highly sensitive holdings information, yet it gives no warning about privacy, retention, transmission, or exposure risks. In a financial context, portfolio composition and watchlist data can reveal wealth, strategy, and intentions, making accidental disclosure materially harmful.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The quick-start instructions encourage users to run monitoring on current holdings without any user-facing warning about the sensitivity of financial portfolio data. Because this skill's context is real-time stock monitoring, the absence of disclosure and safe-handling guidance increases the chance that users unknowingly expose investment positions to logs, remote APIs, or shared environments.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The entire skill file is written in Chinese and does not indicate that language selection is optional or that the content is intentionally limited to a Chinese-speaking or region-specific context. Under the policy rule for natural-language violations, forcing a specific language without user opt-in can be a locale-policy issue.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The document presents explicit buy/sell signals such as golden cross and death cross as actionable trading cues without any risk disclosure, uncertainty framing, or statement that the material is informational only. In a stock-analysis skill, users may reasonably rely on these signals for real trading decisions, which increases the chance of financial harm from overconfident or decontextualized use.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The module description, usage examples, and all user-facing text are written in Chinese, and the script does not provide any option to select another language. This is a natural-language policy concern because it imposes a specific language on users without documented opt-in or justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The script sends derived portfolio information to an external service by constructing a query from locally held stock identifiers. Even though it does not transmit position sizes or cost basis, revealing current holdings to a third-party market-data provider can expose sensitive investment interests without any user notice, consent flow, or privacy warning. In a portfolio-analysis skill, that makes the issue more credible because the code processes real personal financial data from the user's home directory.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The natural-language docstring and all user-facing output are written exclusively in Chinese, with no indication that the user can choose another language or locale. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

SQP-3 applies to all file types and includes language or locale policy violations. The file forces a single language presentation throughout, with no opt-in, alternative language option, or justification that the content is specifically intended for a Chinese-language audience.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The module title, description, usage, and runtime messages are presented only in Chinese. Per the policy, forcing a specific language without user choice can be a natural-language policy violation unless the locale restriction is documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

This code contains natural-language documentation entirely in Chinese, including function docstrings and comments, but does not indicate that the skill is intentionally China-specific or that users may choose another language. Under the policy, forcing a specific language without opt-in or documented justification can be a locale-policy violation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

The script automatically reads a portfolio file from ~/.investment-portfolio/holdings.jsonl without clearly disclosing that behavior in its user-facing description or requiring explicit opt-in. While it does not transmit the file contents directly, it processes potentially sensitive financial holdings data, which creates a privacy risk and violates least-surprise expectations for a local monitoring tool.

Content

No source excerpt is available for this finding.