Security audit

A股实战决策专家

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed A-share trading analysis skill, but users should treat its trading suggestions as high-risk decision support rather than instructions to execute automatically.

Install only if you want a Chinese A-share trading-analysis assistant. Do not let it place trades automatically, review any Buy/Sell or position-sizing output yourself, and require explicit approval before enabling scheduled checks, IM pushes, or online documents containing holdings or trade plans.

A.I.G

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

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (17)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The trigger list for stock-picking includes very broad everyday terms such as '有哪些' and '推荐', which can cause the skill to activate on ordinary conversation that is not intended to invoke financial analysis. In a trading-related skill, accidental invocation is risky because it may prompt market data lookups or investment-style recommendations without clear user intent, increasing the chance of inappropriate or misleading financial guidance.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The real-time quote scenario uses generic triggers like '价格' and '多少钱' without requiring a stock or securities context. This makes accidental activation likely during unrelated conversations, and in a finance skill that can lead to unintended disclosure of quote data, unnecessary tool usage, or responses that appear to be market advice when the user did not clearly request it.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The post-market review scenario uses extremely common words such as '今天' and '收盘' and relies on a time restriction that is described informally rather than enforced as an activation rule. This creates a high risk of inadvertent triggering during ordinary conversation, causing unsolicited market summaries, tool calls, or finance guidance outside the intended context or time window.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The skill instructs the agent to use scheduled tasks and IM/document outputs for holdings monitoring and reminders, but it does not require explicit user consent, disclosure of what data will be sent, or where that data will be stored. Because the skill processes sensitive financial holdings and trading plans, pushing this information to external messaging or document systems can leak private portfolio data or create unauthorized persistent records.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The trigger list is extremely broad and includes generic finance terms such as '股票', '价格', '仓位', and even loose phrasings like asking how a stock is doing. In an agent environment, this can cause the skill to auto-activate for many loosely related conversations, increasing the chance of over-collection of user input, unintended tool use, or routing users into a high-risk financial-decision workflow without clear consent.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The output preference explicitly states "中文为主" (Chinese-first) as a default behavior, but the document does not offer the user a language choice or require opt-in. This is a natural-language locale policy issue because it imposes a language preference globally rather than adapting to user preference.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The document establishes a strict -10% hard stop-loss as a core rule, then later normalizes overriding that rule for a deeply losing position by stating the stop is 'triggered, but not executable' and recommending continued holding. In a trading-discipline skill, this contradiction can encourage users to abandon defined risk controls exactly when losses are largest, increasing exposure to uncontrolled drawdowns and undermining the framework's claimed safety logic.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The entire skill reference is written in Chinese and includes Chinese-specific command/query examples, but it does not state that the skill is region-specific or offer any user opt-in for language preference. Under the language/locale policy rule, a skill that effectively requires a specific language without user choice should be flagged.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This markdown file contains user-facing natural-language guidance entirely in Chinese, and does not indicate that language selection is optional or constrained to a China-specific audience. The policy for this audit requires flagging forced language or locale usage when no user opt-in or documented justification is provided.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The entire skill is written as a prescriptive framework in Chinese for analyst behavior and does not indicate that language is optional or user-selectable. Under the policy rule, forcing a specific language or locale without user opt-in is a natural-language policy violation unless the constraint is explicitly justified.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 87% confidence
Finding: 文件中的标题、触发词、输出模板和操作说明均默认使用中文，没有看到允许用户选择语言或声明该技能仅面向特定中文用户场景的说明。按规则，这可能构成未经用户选择即强制特定语言/locale 的自然语言策略问题。
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The fundamental-analysis triggers include broad finance terms such as '业绩' and '主营' that can appear in general discussion, making activation boundaries unclear. While less immediately dangerous than direct trading triggers, this can still cause the skill to enter an investment-analysis workflow unexpectedly and present authoritative-seeming financial assessments without explicit intent.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The event-driven scenario includes common market and policy words like '政策' and '订单', which are not specific enough to safely trigger a securities workflow. In context, this can lead the system to infer trading intent from ordinary macroeconomic or business discussion and produce speculative catalyst analysis or candidate lists without a clear request.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The portfolio-diagnosis scenario can be triggered by broad terms like '我的票' or '调仓' without requiring actual holdings data or a clear securities context. Because this workflow produces position-sizing and buy/sell adjustment suggestions, accidental activation could result in highly actionable portfolio guidance being generated on incomplete or inferred information.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The document gives concrete, high-risk speculative trading guidance, including screening rules, entry timing, momentum-chasing criteria, and position-control heuristics, but provides no warning about capital loss, suitability, or that the content is not financial advice. In skill form, this can directly influence user behavior toward short-term speculative trades in volatile securities, increasing the chance of substantial financial harm.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: SQP-3 applies to all file types and covers natural-language policy violations such as forcing a specific language without user opt-in. This file presents all operational guidance exclusively in Chinese and contains no indication that the user can select another language or that the locale restriction is justified as region-specific.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 72% confidence
Finding: The document is written entirely in Chinese and is explicitly framed around A-share market behavior and norms, without offering any language choice or clarifying that this locale-specific scope is intentional for a region-specific use case. This can violate language/locale policy when a skill implicitly forces one locale without user opt-in.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.