Security audit

Finance Analysis

Security checks for vulnerabilities and agentic risk

Overview

This finance skill is not destructive or hidden, but it presents stock-specific valuation, risk, and buy-style guidance that is partly hardcoded or under-caveated.

Review carefully before installing or relying on it. Treat valuation, risk score, and buy/sell wording as illustrative unless the publisher replaces hardcoded outputs with current company-specific calculations, documents data sources and assumptions, and adds clear financial-advice and API-token handling warnings.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Intent-Code Divergence

Medium

Confidence: 98% confidence
Finding: The risk_assessment function presents itself as a real stock-specific risk evaluation but never uses the provided stock_code to fetch or compute any data. This can mislead users into making financial decisions based on fabricated, static output, which is especially risky in a finance-analysis skill where users may reasonably trust the results as actionable analysis.

Intent-Code Divergence

Medium

Confidence: 98% confidence
Finding: The valuation_dcf function claims to perform DCF valuation for a supplied stock but only prints fixed example assumptions and results, ignoring the input entirely. In a financial tool, this is dangerous because users may interpret the fabricated valuation as a real security-specific estimate and act on false information.

Intent-Code Divergence

Medium

Confidence: 98% confidence
Finding: The relative valuation function advertises stock-specific comparative valuation but outputs static multiples and a fixed recommendation without performing any analysis. Because the surrounding CLI is framed as a finance-analysis tool, this deceptive behavior can cause users to rely on bogus valuation conclusions.

Missing User Warnings

Low

Confidence: 91% confidence
Finding: The README documents use of a third-party financial data source via Tushare and an optional API token, but it does not warn users that stock identifiers, query parameters, and token-authenticated requests may be transmitted to an external service. This is a real transparency and privacy issue because users may unknowingly send investment-interest metadata or use regulated/internal tokens without understanding the data flow.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The documentation instructs users to provide a TUSHARE_TOKEN credential but gives no warning about sensitive secret handling, storage, or that the skill accesses an external financial data service. This can lead users to expose API tokens in shell history, logs, screenshots, shared terminals, or unsafe environments, increasing the risk of credential leakage and unauthorized API use.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The code produces explicit investment recommendations such as '可以买入', '值得买入', and '强烈买入' based on a simplistic heuristic, without any disclaimer that the output is illustrative and not financial advice. In a finance-analysis skill, users may reasonably rely on these recommendations for real decisions, creating consumer-harm, compliance, and trust risks if the guidance is misleading or unsuitable.

Static analysis

No suspicious patterns detected.