ClawHub
Back to skill

Security audit

Polymarket Odds

Security checks across malware telemetry and agentic risk

Overview

This is a read-only Polymarket lookup skill that contacts a disclosed public API and does not access credentials, persist, or modify data.

Install only if you are comfortable with the agent making unauthenticated requests to Polymarket's public API when market odds are requested. Treat returned prices as market data, not financial advice, and prefer explicit Polymarket-related prompts to avoid unnecessary lookups.

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
95% confidence
Finding
The skill performs network access to Polymarket's public API, but the manifest does not declare any corresponding permission or capability. Undeclared network behavior weakens security review and user/operator visibility, and can become dangerous if the script is modified to contact additional hosts or exfiltrate data under the guise of a market query tool.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The description says to use the skill for questions about prediction markets, betting odds, market prices, event probabilities, or when the user asks about Polymarket data, which is broad enough to capture many ordinary finance, politics, or probability-related requests. Overly broad routing can cause the agent to invoke networked tooling unnecessarily, exposing user queries to external services and increasing the chance of incorrect tool selection or unintended data disclosure.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.