Back to skill

Security audit

Prediction Market Pro Trader

Security checks for vulnerabilities and agentic risk

Overview

This skill is a prediction-market trading advisor that gives explicit trade and position-size recommendations while also saying it is only educational and not financial advice.

Review this skill carefully before installing. It does not appear to contain code, persistence, credential access, or exfiltration behavior, but it is designed to produce actionable prediction-market trading calls and position sizes. Treat outputs as high-risk financial opinions, verify platform legality in your jurisdiction, and do not let an agent place trades or size positions automatically based on this skill.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The disclaimer says the skill is for analysis and education only, but the rest of the document operationalizes concrete market recommendations and paid advisory-style services. In context, this is dangerous because the domain is financial trading, where users may rely on the output to place real-money bets despite the disclaimer.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill presents itself as not providing financial advice, yet its output format explicitly instructs the agent to issue BUY/HOLD/PASS recommendations, trade direction, and position sizing. That contradiction can mislead users and downstream systems into treating regulated trading guidance as mere education, increasing legal/compliance risk and the chance of harmful financial decisions.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The skill gives concrete recommendation formats tied to named trading platforms, but platform-specific legal and residency limitations are not surfaced at the recommendation point. In a prediction-market context, that omission can cause users to act on advice for platforms they cannot legally access or use, creating compliance and consumer-risk issues.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.