Polymarket Analysis

• Outcome: Pass
• Latest audit: Jul 10, 2026 · 2:37 AM UTC
• Version: 2.1.0

Overview

This is a disclosed Polymarket research skill with optional monitoring alerts, not a trading or credential-stealing tool.

Install only if you want read-only Polymarket market and wallet analysis. Before enabling Monitor mode, confirm the market, cron interval, state-file location, delivery channel, and recipient, and remove the cron job and stored state when you no longer want alerts.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Rogue Agent: Self-Modification, Session Persistence
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (7)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 84% confidence
• Finding: The skill advertises and references executable scripts that perform network access and likely monitoring/output behavior, but the skill metadata declares no permissions. This creates a trust and sandboxing gap: a user or host system may treat the skill as low-risk while it can still initiate external requests and potentially write files/logs during monitoring workflows.

Description-Behavior Mismatch

Medium

• Confidence: 96% confidence
• Finding: The document instructs creation of persistent cron jobs and delivery of alerts, which goes beyond a read-only 'analysis only, no execution' framing in the skill metadata. This mismatch can cause users or reviewers to underestimate that the skill triggers ongoing automated actions and notifications after the initial interaction.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The setup describes reading and updating a persistent state file on disk, contradicting a read-only analysis expectation. Hidden statefulness can create privacy, integrity, and auditability issues because future behavior depends on stored data the user may not realize exists.

Missing User Warnings

Low

• Confidence: 88% confidence
• Finding: The skill documents writing market-monitoring state to `.claude/skills/polymarket-analysis/state/{market-id}.json` without warning the user that local data will be stored. Even if the data is not highly sensitive, undisclosed storage can expose user activity patterns and create unexpected residue on shared or managed systems.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The documentation configures ongoing alert delivery to WhatsApp, Telegram, and Discord without explicit disclosure that monitoring information may be transmitted to third-party services. This can leak trading interests, market focus, timing, or behavioral metadata off-platform and outside the user's expected trust boundary.

Missing User Warnings

Low

• Confidence: 88% confidence
• Finding: The document explicitly promotes monitoring identifiable wallets, querying their positions, and generating alerts about their trading activity without any discussion of privacy, consent, legal restrictions, or anti-harassment safeguards. While blockchain data may be public, packaging it into a whale-tracking workflow lowers the barrier for targeted surveillance and copy-trading behavior, which creates privacy and misuse concerns in this context.

Session Persistence

Medium

• Category: Rogue Agent

Content

On each cron run, the agent:

1. **Fetch current data** via Gamma API
2. **Load previous state** from state file
3. **Compare thresholds:**

| Check | Condition | Alert |

• Confidence: 84% confidence
• Finding: Load previous state

VirusTotal

66/66 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.