Back to skill

Security audit

Polymarket Arbitrage Pro

Security checks for vulnerabilities and agentic risk

Overview

This skill is a high-risk Polymarket trading bot that asks for a wallet private key and is designed to place recurring live orders without clear confirmations, limits, or safe default controls.

Review this carefully before installing. Do not use a primary wallet private key; if testing, use a dedicated low-balance wallet and assume any configured funds could be traded automatically. Treat scan/start as potentially live-trading workflows, and avoid use unless you are comfortable with the lack of confirmations, limits, and reliable arbitrage safeguards.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (3)

T09 · Insecure Skill Coding Practices

Error
Location
polymarket-trade.js:291
Finding

Automatic Trading Logic Executes Unhedged Speculative Orders Instead of Arbitrage

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
polymarket-trade.js:172
Finding

Billing API Key and Wallet Identifier Are Transmitted to an External Service

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
polymarket-trade.js:187
Finding

Billing Enforcement Fails Open and Sends a Zero-Amount Charge

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (15)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill description markets 'automatic detection and execution of arbitrage trades' and states that each invocation auto-charges fees, but it does not clearly foreground that the skill may place real-money trades without per-trade user confirmation. In a financial trading context, silent or background execution materially raises the chance of unintended orders, loss of funds, and abuse if the skill is misconfigured or behaves unexpectedly.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The README explicitly instructs users to export and provide a raw Polygon wallet private key via an environment variable for automated trading. Requesting direct custody secrets in a third-party skill materially increases wallet-compromise risk, especially because the skill is designed to execute financial transactions and gives only minimal generic trading-risk warnings rather than strong credential-handling warnings.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The documented behavior understates several highly sensitive actions: use of a raw private key for signing, communication with an external billing service, and continuous monitoring or polling. This mismatch is especially dangerous because users may believe they are invoking a simple arbitrage helper while actually granting a skill the ability to spend crypto assets, incur charges, and run persistent automated activity.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The scan command can place real market orders automatically whenever a wallet is configured, without any explicit user confirmation, dry-run default, max-loss guard, or final warning. Because this directly triggers live financial transactions from routine discovery logic, accidental invocation or misleading integration could cause immediate unwanted spending and market exposure.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The start command enables unattended recurring scans every 60 seconds, and those scans may submit live orders automatically if a wallet is present. This compounds the previous issue by allowing repeated real-money trades with no operator approval per cycle, no circuit breaker, and no visible risk acknowledgement, increasing the chance of runaway losses or abuse.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill advertises capabilities that require environment access and network communication, but it declares no explicit tool scope or permissions. In a security-sensitive context involving wallet keys and automated trading, missing scope declarations hide the real trust boundary and can cause users or platforms to authorize a skill without understanding that it can access secrets and interact with external services.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill explicitly instructs users to export and provide a raw Polygon private key for automated trading, without strong warnings against key reuse, hot-wallet risk, or safer alternatives. Because a raw private key grants full control over wallet funds, any compromise, misuse, or hidden transaction logic could result in irreversible theft of crypto assets.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

User-facing descriptions, logs, help text, and warnings are written in Chinese throughout the file, with no indication that the skill is region-specific or that users may choose another language. This can violate language/locale policy where skills must not force a specific language without user opt-in.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The code does not enforce paid access before executing the scan/trading flow. If SKILLPAY_KEY is missing or the charge request errors, chargeUser() returns success and the program continues, allowing use of the skill contrary to the stated billing model. In a paid trading skill this is a real integrity/business-logic flaw, though not as severe as direct fund theft.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
75% confidence
Finding

The visible user-facing description and instructions are entirely in Chinese, and the file does not offer an alternative language or state that the skill is region-specific. Under the language/locale policy, forcing a specific language without user opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest description is written only in Chinese ("Polymarket预测市场套利工具"), which imposes a specific language presentation without any indication of user opt-in or locale-specific justification. This matches the policy category for language or locale constraints expressed in natural language.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
89% confidence
Finding

Using caret ranges for dependencies allows newer package versions to be installed without explicit review, which can introduce supply-chain risk, behavioral changes, or inadvertently pull in a vulnerable release. In a trading/arbitrage skill that may handle API calls and blockchain interactions, unexpected dependency changes can have outsized financial impact.

Content

Scanner excerpt · package.json (reported line 15)May include surrounding context.

json
"node": ">=14"
  },
  "dependencies": {
    "node-fetch": "^2.7.0",
    "ethers": "^5.7.2"
  }
}

Unverifiable Dependency: node-fetch has 3 known advisory(ies) (CVE-2022-0235 (node-fetch forwards secure headers to untrusted sites); CVE-2022-2596 (node-fetch Inefficient Regular Expression Complexity ); CVE-2020-15168 (The `size` option isn't honored after following a redirect in node-fetch)), but the manifest does not pin a version, so it is unknown whether the installed release is affected

Low
Category
Supply Chain
Confidence
83% confidence
Finding

The manifest references node-fetch without pinning an exact version, while known advisories exist for some releases, making it impossible to verify from this file alone whether deployments are affected. In a tool that likely performs network requests for market data or execution, vulnerable HTTP client behavior could expose sensitive headers, enable denial-of-service conditions, or weaken request safety.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
89% confidence
Finding

Using a non-exact version for ethers permits automatic uptake of newer compatible releases, increasing supply-chain risk and the chance of unexpected runtime or security-impacting changes. Because this skill is described as executing arbitrage trades and charging per invocation, dependency drift in a blockchain library can directly affect transaction construction, signing, or fund handling.

Content

Scanner excerpt · package.json (reported line 16)May include surrounding context.

json
},
  "dependencies": {
    "node-fetch": "^2.7.0",
    "ethers": "^5.7.2"
  }
}

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The file header and CLI help both state that POLYMARKET_PRIVATE_KEY is required. In practice, initWallet only logs a warning and returns false, and scan can still execute market scanning and billing-related network calls without a wallet; only order placement is gated.

Content

No source excerpt is available for this finding.