ClawHub

Crypto Market

Security checks across malware telemetry and agentic risk

Overview

This is a crypto price-monitoring skill whose behavior matches its purpose, with documentation notes users should read before use.

Specify --exchange if it matters which exchange receives requests, because the code currently defaults to OKX despite documentation saying Binance. Do not add exchange API keys for simple price monitoring, and avoid committing any credentials if adapting the reference example. Review or delete ~/.config/crypto/alerts.json if you no longer want local alert rules stored.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding
The documented behavior does not fully match the implemented behavior: local persistence is omitted from the high-level description, and the default exchange reportedly differs from the stated default. This can mislead users about where data is stored and which external service receives requests, undermining informed consent and creating operational/security risk if users trust the documentation.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The document includes a private API credential example with hardcoded placeholder fields for apiKey and secret but does not warn readers against committing secrets, exposing them in logs, or storing them insecurely. In a crypto-trading context, users may copy this pattern directly into scripts or configs, and leaked exchange API keys can enable account reconnaissance or trading actions depending on permissions.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal