Security audit

虾皮市场温度指标分析

Security checks across malware telemetry and agentic risk

Overview

This is a coherent market-analysis skill, but users should handle API tokens carefully and treat its trading outputs as informational rather than personal financial advice.

Install only if you are comfortable using a third-party DaxiAPI token and reviewing financial outputs critically. Prefer Authorization bearer headers, avoid token-in-URL examples, avoid putting real tokens in shared shell history or synced dotfiles, and treat any buy/sell language as non-personalized analysis rather than instructions to trade.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (6)

Intent-Code Divergence

Medium

Confidence: 94% confidence
Finding: The document presents passing the API token as a URL query parameter while elsewhere stating that putting tokens in URLs is unsafe. Query-string tokens are commonly exposed via browser history, logs, reverse proxies, analytics, and Referer headers, so including this pattern in official documentation can directly lead users to leak credentials.

Missing User Warnings

Medium

Confidence: 86% confidence
Finding: The documentation instructs users to set an API token directly on the command line without warning about credential handling. Tokens entered this way may be exposed through shell history, terminal logging, process inspection, screenshots, or shared command transcripts, which can lead to credential leakage and unauthorized API use.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: This section gives concrete action-oriented investment guidance such as suggesting entry timing and staged selling before any prominent non-advisory or risk warning appears. In a financial-analysis skill, users may reasonably treat these statements as actionable advice, increasing the chance of unsuitable trading decisions and user harm.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The buy-signal table escalates from observation to 'strongly recommend phased buying or DCA' without an adjacent warning about losses, uncertainty, or user suitability. This is especially risky because a structured signal table can be interpreted as a decision rule, encouraging users to act mechanically on market indicators.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The '估值底部' section instructs users to ignore emotion indicators, buy more as prices fall, and calls valuation bottoms the best long-term entry opportunity. That combination is high-conviction financial guidance that can amplify losses in prolonged downtrends, especially because it downplays conflicting risk signals.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The guide tells users to append an API token directly into shell startup files like ~/.zshrc, which stores a long-lived secret in plaintext on disk and may expose it to local compromise, backups, screen sharing, dotfile sync, or accidental disclosure. Although this is common setup guidance, the document does not clearly warn about the security tradeoff of persisting credentials this way or recommend safer secret-management alternatives.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.