Security audit

宝总霸盘Skill — 全市场智能操盘系统

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed, instruction-only stock analysis skill with broad finance triggers, but it does not install code, access accounts, place trades, or hide behavior.

Install this only if you want a Chinese-language stock analysis and trading-decision framework that may activate broadly during market conversations. Verify market data independently, do not connect brokerage credentials unless you fully trust the separate tool handling them, and treat any buy, sell, stop-loss, or portfolio guidance as advisory rather than automatic action.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The skill’s activation description and keyword list are extremely broad, covering common investing and market-discussion phrases that can appear in ordinary conversations. This can cause unintended invocation, making the assistant apply the skill in contexts where the user did not explicitly request stock-trading analysis, which is especially risky because the skill produces actionable financial guidance.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The trigger examples include vague phrases such as asking what opportunities exist today, which are normal conversational expressions and not reliable indicators that the user wants this specific skill engaged. In a finance context, accidental activation can steer a general chat into investment recommendations or monitoring behavior without clear consent.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The execution-layer triggers like requests to set reminders or monitor a stock are generic and could match many benign assistant tasks, causing the skill to activate too easily. Because this skill includes trading, monitoring, and risk-control workflows, such misfires could result in unintended financial analysis or persistent watchlist behavior.

Vague Triggers

Low

Confidence: 82% confidence
Finding: The risk-check triggers contain generic phrases like asking how an account is doing, which are ambiguous and may refer to non-financial accounts or general status checks. While lower severity than direct trading prompts, this can still route unrelated conversations into a financial risk-analysis flow.

Natural-Language Policy Violations

Medium

Confidence: 84% confidence
Finding: The skill description is written to operate in a Chinese-language interaction context without indicating language negotiation or user choice. This can create usability and safety issues by overriding the user’s preferred language, increasing the chance of misunderstanding financial analysis, risk disclosures, or trading-related recommendations.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.