ClawHub
Back to skill

Security audit

加密货币监控与告警

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed crypto price-monitoring and alert helper that uses public market APIs and local alert storage, with one advisory-report wording issue users should treat cautiously.

Before installing, be aware that the skill may call public crypto-price APIs and can save alert rules locally. Treat any market report language as informational only, especially the template section labeled as operation suggestions; do not rely on it as financial advice or as a trading signal.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The daily report template includes sections for '操作建议' and free-form 'analysis', which can steer users toward investment decisions despite the skill metadata explicitly limiting the tool to monitoring only and not providing investment advice. In a crypto context, this mismatch increases the chance that downstream agents or users will treat generated output as actionable financial guidance.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The trigger phrases are broad enough to match generic conversation about prices, reports, or market activity, which can cause the skill to activate when the user did not intend to use it. Mis-triggering is mainly a safety and UX issue here, but in a network-enabled skill it can still cause unintended external requests or creation of alerts based on ambiguous input.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.