Autonomous Trading System

• Outcome: Review
• Latest audit: May 29, 2026 at 9:27 PM UTC
• Version:
  1.0.1

Overview

This skill describes an unattended crypto trading bot that may use Binance credentials to place live trades, but its safeguards and reviewable implementation are not sufficient for that level of financial authority.

Review carefully before installing. Do not connect this to a real exchange account as-is; require the missing source code and dependencies, start with paper trading or Binance testnet, and only use dedicated trade-only API keys with withdrawals disabled, IP restrictions, strict limits, alerts, and an emergency stop.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Missing User Warnings

High

• Confidence: 94% confidence
• Finding: The document repeatedly presents a 'fully automated', 'zero intervention' trading system that can make and execute decisions on user assets, but it does not include any explicit user-facing warning about financial loss, irreversible order execution, or the need for informed consent. In the context of an agent skill, this omission is dangerous because it normalizes autonomous high-risk financial behavior and could cause users to enable trading without understanding the consequences.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The architecture explicitly advertises autonomous decision-making with no per-trade human confirmation, which materially increases risk because the system can act immediately on flawed signals, bad data, API failures, or unsafe self-adjustments. In a trading skill, removing human approval without a corresponding warning, consent model, or default safety gate makes unintended loss of funds much more likely.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The order management section documents use of market orders, stop-market orders, and automatic closing logic, but does not warn that these can execute at unfavorable prices, trigger during volatility, and create irreversible losses through slippage or gap risk. Because this is an autonomous execution system, the absence of such warnings reduces operator awareness of a core operational hazard.

VirusTotal

63/63 vendors flagged this skill as clean.