openclaw-polymarket-trading-bot

Warn

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to clone and execute code from a non-trusted GitHub repository (Golrypavium/openclaw-polymarket-trading-bot.git), which introduces a significant supply chain risk.
  • [INDIRECT_PROMPT_INJECTION]: The bot ingests external market data used to construct prompts for an LLM that makes financial trading decisions, which could be manipulated to influence the bot's behavior.
  • Ingestion points: Market price ticks and whale transaction data are fetched from the Polymarket API via core/adapters/eventSource.ts.
  • Boundary markers: No explicit delimiters or warnings to ignore malicious instructions are present in the prompt generation logic within core/ai/remoteScorer.ts.
  • Capability inventory: The bot possesses capabilities for network requests (fetch), local file system writes (fs.writeFileSync), and executing financial transactions via the Polymarket CLOB API.
  • Sanitization: The data is processed through numeric formatting (toFixed) and parsing (parseFloat), but lacks robust validation against adversarial injection.
  • [COMMAND_EXECUTION]: The installation process involves executing shell commands such as npm install and npm start on downloaded third-party code.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 12, 2026, 06:50 PM
Security Audit — agent-trust-hub — openclaw-polymarket-trading-bot