Back to skill

Security audit

futu-trading-bot

Security checks across malware telemetry and agentic risk

Overview

This trading skill is mostly transparent, but it exposes a direct path that could bypass its confirmation checks for live brokerage actions.

Install only if you are comfortable giving an agent access to Futu OpenD and potential live brokerage operations. Use SIMULATE by default, require explicit confirmation for live actions, keep credentials in a private config file, and avoid letting agents import or call internal trade_service objects directly.

Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • 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)

Intent-Code Divergence

Medium
Confidence
91% confidence
Finding
The module-level _trade_service instance remains directly importable, despite the comment claiming the internal service is not exposed to prevent bypassing validation. Any consumer of the skill can call _trade_service.place_order(), modify_order(), or cancel_all_orders() directly and avoid the confirm=True safety gates meant to protect REAL trades and bulk cancellation.

Session Persistence

Medium
Category
Rogue Agent
Content
signal.signal(signal.SIGTERM, handle_exit)
signal.signal(signal.SIGINT, handle_exit)

# Write PID file
with open(PID_FILE, "w") as f:
    f.write(str(os.getpid()))
Confidence
77% confidence
Finding
The provided strategy template writes a PID file and log file for a long-running process, which creates persistence and process-control state on disk. While intended for legitimate strategy management, stale or attacker-manipulated PID/log paths could enable accidental interference with other processes, unwanted persistence, or file clobbering if unsafe paths are later substituted.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.