Back to skill

Security audit

Binance DCA Test

Security checks across malware telemetry and agentic risk

Overview

This skill is not clearly malicious, but it can steer an agent into live Binance crypto purchases and recurring trading without enough guardrails.

Install only if you intentionally want an agent-assisted Binance trading workflow. Before live use, inspect the missing dca.sh implementation, start on Binance testnet, use a restricted spot-trading API key with withdrawals disabled, set explicit spending limits, and require manual confirmation before every real order or recurring schedule.

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
84% confidence
Finding
The skill description is broad enough to trigger on generic investing or buying requests, which can cause the agent to invoke a real-trading capability in contexts where the user did not explicitly request Binance order execution. Because this skill can place live buy orders, overbroad routing materially increases the chance of unintended financial actions or escalation from informational advice to execution.

Missing User Warnings

High
Confidence
94% confidence
Finding
The documentation provides direct examples for placing live market and limit buy orders but does not present a prominent warning that these actions can spend real funds immediately. In a skill that interacts with exchange APIs, missing or weak warnings increase the risk of accidental live trades, especially if users or agents copy quick-start commands without realizing they are not confined to testnet.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.