Polymarket Scanner

• Outcome: Review
• Latest audit: May 29, 2026 at 9:22 PM UTC
• Version:
  1.0.0

Overview

This is a coherent Polymarket trading skill, but it can submit trades by default and does not implement the promised real-money confirmation or several stated risk controls.

Review before installing. Use only a virtual or tightly capped Simmer account unless you independently verify that the service enforces real-money confirmation. Do not provide a live trading key or schedule this with cron until execution requires explicit opt-in and the advertised daily limits, stop-losses, and per-trade confirmations are actually implemented.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 93% confidence
• Finding: The skill advertises capabilities that involve environment-variable access, network access, and file writing, but it does not declare permissions or boundaries for those actions. In a trading skill, this is dangerous because it can access API keys, place external requests, and persist potentially sensitive trading data without clear governance or user visibility.

Vague Triggers

Medium

• Confidence: 83% confidence
• Finding: The trigger list includes broad terms like "trade," "market scan," and "arbitrage opportunities," which can activate the skill for requests outside the user's intended context. Because this skill can lead to financial actions, overly broad routing increases the chance of accidental invocation and unsafe trade-related responses in unrelated conversations.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The description explicitly allows execution of real-money USDC trades, but it does not present a prominent upfront warning about financial risk, irreversible market actions, or the possibility of loss. In the context of an automated trading skill, this materially raises harm because users may invoke it without understanding that live funds could be used.

Missing User Warnings

High

• Confidence: 99% confidence
• Finding: The script can automatically place trades in the default execution path with no interactive confirmation, dry-run default, or hard stop when real trading is enabled. In a skill explicitly designed for scanning and trading prediction markets, this is especially dangerous because a routine invocation can commit real funds based solely on heuristic thresholds and external API responses.

VirusTotal

65/65 vendors flagged this skill as clean.