Polymarket MCP Server
Enables AI agents to interact with the Polymarket prediction market platform to discover markets, analyze real-time pricing, and monitor public portfolio data. It provides comprehensive access to market analytics, implied probabilities, and order books through the Model Context Protocol.
README
📊 Polymarket MCP Server
📖 Overview
The Polymarket MCP Server enables AI agents to interact with Polymarket, a prediction market platform on Polygon. This server provides read-only access to market data (Gamma API), real-time pricing and order books (CLOB public API), public user data (Data API), comments/profiles, and bridge quotes/status.
By implementing the Model Context Protocol (MCP), this server allows Large Language Models (LLMs) to discover prediction markets, analyze odds (probabilities), and monitor positions or activity from public data sources.
✨ Features
- Market Discovery: Search and filter prediction markets by keywords, tags, status, and events.
- Real-time Pricing: Access live price data, implied probabilities, and order books for any outcome token.
- Public Portfolio Data: Monitor positions, trade history, and activity for specific wallet addresses (public Data API).
- Comments & Profiles: Retrieve public profiles and market/event comments.
- Realtime Subscriptions: Subscribe to price, orderbook, and resolution updates.
- Bridge Utilities: Fetch supported assets and bridge quotes/status.
📦 Installation
🔧 Build from Source
git clone https://github.com/traderalvin1/polymarket-mcp.git
cd polymarket-mcp
pnpm install
pnpm run build
⚡ Running with an MCP Client
Add the following configuration to your MCP client settings (e.g., claude_desktop_config.json).
📋 Minimal Configuration (Read-Only)
{
"mcpServers": {
"polymarket": {
"command": "node",
"args": ["/path/to/polymarket-mcp/dist/index.js"]
}
}
}
🔐 Configuration (Environment Variables)
| Variable | Required | Description | Default |
|---|---|---|---|
CLOB_API_BASE |
No | Polymarket CLOB API base URL | https://clob.polymarket.com |
CHAIN_ID |
No | Blockchain network chain ID | 137 (Polygon) |
GAMMA_API_URL |
No | Polymarket Gamma API base URL | https://gamma-api.polymarket.com |
DATA_API_URL |
No | Polymarket Data API base URL | https://data-api.polymarket.com |
BRIDGE_API_URL |
No | Polymarket Bridge API base URL | https://bridge.polymarket.com |
💡 Usage Examples
🔍 Market Discovery
- "What are the most active prediction markets on Polymarket right now?"
- "Search for markets related to 'Bitcoin' or 'BTC'."
- "Find markets in the 'Crypto' category."
- "What events are trending on Polymarket today?"
📊 Analytics & Pricing
- "Show me the order book for the 2024 election market."
- "What is the current probability implied by the price of the 'Yes' token?"
- "Get detailed information about the 'will-trump-win-2024' market."
💼 Portfolio & Trading (Requires Private Key)
- "What's my current USDC balance and allowance?"
- "Show me all my open orders across all markets."
- "Place a buy order for 100 shares at 0.65 price."
- "Cancel all my open orders on this market."
🛠️ MCP Tools
Full tool documentation is maintained in:
TOOLS_REFERENCE.md— complete tool list, parameters, dependencies, edge casesUSAGE_EXAMPLES.md— workflows and example calls
For agents, call get_tool_guide to get a concise, runtime usage guide.
🙏 Credits
This project is based on and modified from:
- https://github.com/IQAIcom/mcp-polymarket
👨💻 Development
🏗️ Build Project
pnpm run build
👁️ Development Mode (Watch)
pnpm run watch
✅ Linting & Formatting
pnpm run lint
pnpm run format
📁 Project Structure
src/tools/: Individual tool definitionssrc/services/: API client and business logicsrc/index.ts: Server entry point
📚 Resources
⚠️ Disclaimer
This project is an unofficial tool and is not directly affiliated with Polymarket. It interacts with financial and prediction market data. Users should exercise caution and verify all data independently. Trading in prediction markets involves risk.
📄 License
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。