Security audit

对上市公司进行系统性投资价值分析，支持 A 股、港股、美股

Security checks across malware telemetry and agentic risk

Overview

This stock-analysis skill matches its stated purpose, but users should review it carefully because it starts an unauthenticated background report server and can produce actionable investment guidance.

Install only if you are comfortable with a finance skill that fetches public market data, writes cached results, and can run a local report server. Use the report feature only where port 8888 is not exposed beyond your machine or container, stop the background server when finished, and treat all ratings, target prices, stop-losses, and allocation percentages as educational analysis rather than professional financial advice.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (5)

Vague Triggers

Medium

Confidence: 85% confidence
Finding: The phrase '分析请求触发' is broad and does not clearly define what user action or confirmation is required before the skill automatically invokes multiple external scripts. In an agent setting, ambiguous auto-execution conditions can cause unintended network access and tool use, increasing the risk of over-collection, privacy issues, or abuse through prompt-triggered actions.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The trigger example '同行业哪个更好' is broad conversational language that could easily appear in ordinary investing discussions, increasing the chance the skill is invoked unintentionally. In a finance context, accidental activation is more concerning because it can steer users toward investment recommendations or portfolio allocation suggestions without an explicit, deliberate request to use this specific skill.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill explicitly documents starting an HTTP server on port 8888 and exposing a direct download URL containing the server IP, but provides no warning that the report may become reachable from other hosts on the local network or beyond depending on binding and firewall configuration. Because the generated HTML report likely contains financial analysis content and possibly user-requested company context, this creates an unnecessary local data exposure surface and normalizes network service exposure from a skill.

Missing User Warnings

Low

Confidence: 95% confidence
Finding: The script sends the provided stock code to an external third-party service over outbound network requests without any disclosure, consent, or privacy notice. In addition, it uses plain HTTP rather than HTTPS, which exposes requested identifiers and returned data to interception or tampering on the network path, making the issue more significant than a simple documentation gap.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: This template includes concrete portfolio allocation ranges, trading actions, and stop-loss guidance presented as actionable investment advice, while only ending with a minimal generic disclaimer. In an agent skill context, this can cause the system to generate personalized-seeming financial recommendations without sufficient risk warnings, suitability checks, or jurisdictional compliance safeguards.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.