ClawHub
Back to skill

Security audit

Smart Agent Skill

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Polymarket trading skill, but it can run every 15 minutes and place real-money trades if live mode is enabled.

Review before installing. Use a limited or test AION key if available, leave RUN_LIVE unset unless you intentionally want real trades, keep trade sizes small, and monitor the scheduled automaton because live mode can place repeated trades every 15 minutes without asking again.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill documentation indicates use of the AION_API_KEY environment variable, but the skill declares no permissions despite requiring environment access. This creates a transparency and policy gap: an installer or reviewer may not realize the skill needs secret access, and a runtime may grant broader access than explicitly intended. In a trading skill, hidden or undeclared env usage is more sensitive because the referenced credential can enable market reads and potentially live trade execution.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The manifest configures automatic execution on a broad 15-minute cron schedule and the skill explicitly supports live trade execution. In a trading context, unattended periodic invocation can repeatedly place or propose trades without a strong external approval gate, increasing the risk of unintended financial loss, runaway automation, or abuse if configuration is unsafe.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.