graph-polymarket-mcp
An MCP server that enables querying Polymarket prediction market data through The Graph subgraphs. It provides tools for accessing market stats, trader P\&L, user positions, and orderbook activity using AI agents.
README
graph-polymarket-mcp
MCP server for querying Polymarket prediction market data via The Graph subgraphs.
Exposes 9 tools that AI agents (Claude, Cursor, etc.) can use to query market data, trader P&L, positions, activity, and orderbook trades.
Published to the MCP Registry as
io.github.PaulieB14/graph-polymarket-mcp
Prerequisites
You need a free Graph API key (takes ~2 minutes):
- Go to The Graph Studio
- Connect your wallet (MetaMask, WalletConnect, etc.)
- Click "API Keys" in the sidebar and create one
- Free tier includes 100,000 queries/month
Installation
npm install -g graph-polymarket-mcp
Or use directly with npx:
npx graph-polymarket-mcp
Configuration
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"graph-polymarket": {
"command": "npx",
"args": ["-y", "graph-polymarket-mcp"],
"env": {
"GRAPH_API_KEY": "your-api-key-here"
}
}
}
}
Claude Code
claude mcp add graph-polymarket -- npx -y graph-polymarket-mcp
Set the environment variable GRAPH_API_KEY before running.
Cursor / Other MCP Clients
Use the stdio transport with npx graph-polymarket-mcp as the command, passing GRAPH_API_KEY as an environment variable.
Available Tools
Core Tools
| Tool | Description |
|---|---|
list_subgraphs |
List all available Polymarket subgraphs with descriptions and key entities |
get_subgraph_schema |
Get the full GraphQL schema for a specific subgraph |
query_subgraph |
Execute a custom GraphQL query against any subgraph |
Domain-Specific Tools
| Tool | Description | Subgraph |
|---|---|---|
get_market_data |
Get market/condition data with outcomes and volumes | Main |
get_account_pnl |
Get a trader's P&L and performance metrics | Beefy P&L |
get_user_positions |
Get a user's current positions | Slimmed P&L |
get_recent_activity |
Get recent splits, merges, and redemptions | Activity |
get_orderbook_trades |
Get recent order fills | Orderbook |
get_global_stats |
Get aggregate platform statistics | Main |
Subgraphs
| Name | IPFS Hash | Description |
|---|---|---|
| Main | QmdyCguLEisTtQFveEkvMhTH7UzjyhnrF9kpvhYeG4QX8a |
Complete ecosystem data |
| Beefy P&L | QmbHwcGkumWdyTK2jYWXV3vX4WyinftEGbuwi7hDkhPWqG |
Comprehensive P&L tracking |
| Slimmed P&L | QmZAYiMeZiWC7ZjdWepek7hy1jbcW3ngimBF9ibTiTtwQU |
Minimal position data |
| Activity | Qmf3qPUsfQ8et6E3QNBmuXXKqUJi91mo5zbsaTkQrSnMAP |
Position management events |
| Orderbook | QmVGA9vvNZtEquVzDpw8wnTFDxVjB6mavTRMTrKuUBhi4t |
Order fill analytics |
Example Queries
Once connected, an AI agent can:
- "What are the current Polymarket global stats?"
- "Show me the latest 20 orderbook trades"
- "What are the positions for address 0x...?"
- "Get the P&L for trader 0x...?"
- "Query the main subgraph for all conditions with more than 100 trades"
Development
git clone https://github.com/PaulieB14/graph-polymarket-mcp.git
cd graph-polymarket-mcp
npm install
npm run build
GRAPH_API_KEY=your-key node build/index.js
License
MIT
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。