ClawHub
Back to skill

Security audit

Crypto Trading Bot

Security checks across malware telemetry and agentic risk

Overview

This skill appears aimed at building crypto trading bots, but it under-discloses important financial and API-key risks for an automated trading workflow.

Review this skill carefully before installing. Use it only for deliberate crypto trading bot work, start with paper or sandbox trading, never provide seed phrases or unrestricted exchange keys, disable withdrawal permissions, and require explicit human review before any live trading or deployment.

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 (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger phrase uses a very broad natural-language request ('整交易Bot') that can match ordinary conversation and unintentionally activate the skill outside a clearly scoped consent flow. In a trading context, accidental activation is more dangerous than for ordinary content because the skill guides users toward automated financial systems and API-connected trading workflows.

Missing User Warnings

High
Confidence
96% confidence
Finding
The skill promotes automated trading, exchange API integration, arbitrage, and flash-loan-related workflows without any visible warnings about financial risk, liquidation risk, exchange policy compliance, credential handling, or the consequences of granting API access. In this context, omission of safeguards is especially dangerous because users may expose API keys or deploy high-risk automation that can rapidly lose funds or execute unintended trades.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal