Questrade MCP Server
An unofficial MCP server that integrates with the Questrade API to provide access to trading accounts, market data, and portfolio information. It enables users to view balances, track positions, search symbols, and analyze market trends through natural language.
README
Questrade MCP Server
An unofficial Model Context Protocol (MCP) server for integrating with the Questrade API, providing access to trading accounts, market data, and portfolio information.
⚠️ Disclaimer: This is an unofficial, community-built integration and is not affiliated with, endorsed by, or supported by Questrade Inc. Use at your own risk.
Features
- 🔐 Authentication: OAuth 2.0 token management with automatic refresh
- 📊 Account Data: Access accounts, positions, balances, and order history
- 📈 Market Data: Real-time quotes, symbol search, and historical candles
- 🛡️ Error Handling: Comprehensive error handling and logging
- 🔧 TypeScript: Full TypeScript support with proper type definitions
Installation
Option 1: Install from npm (Recommended)
npm install -g questrade-mcp-server
Option 2: Clone and Build
-
Clone this repository
-
Install dependencies:
npm install -
Copy the environment template:
cp .env.example .env -
Configure your Questrade API credentials in
.env:QUESTRADE_API_URL=https://api01.iq.questrade.com QUESTRADE_REFRESH_TOKEN=your_refresh_token_here # QUESTRADE_TOKEN_DIR=/path/to/custom/directory
Getting Questrade API Credentials
For detailed information about Questrade's API authorization, see the official API documentation.
Step 1: Generate API Token
-
Log in to your Questrade account or navigate directly to https://apphub.questrade.com/UI/UserApps.aspx
-
In the top right corner, select "API centre" from the drop-down menu under your login name

-
Click "Activate API" and agree to the API access agreement
-
Click "Generate new token" for manual authorization

-
Copy the refresh token provided

Step 2: Configure Environment
-
Copy your refresh token to
.env:QUESTRADE_REFRESH_TOKEN=your_refresh_token_here -
The MCP server will automatically:
- Use your refresh token to get an access token
- Discover the correct API server URL
- Handle token refresh when needed
- Persist new tokens to
~/.questrade-mcp/tokens.json(or system temp directory as fallback)
Important: Refresh tokens are single-use only. The server will attempt to persist new refresh tokens to
~/.questrade-mcp/tokens.json(configurable viaQUESTRADE_TOKEN_DIRenvironment variable), but if a token expires or gets used by another process, you'll need to manually generate a new one following the steps above.
Step 3: Test Your Setup
Verify your token works correctly:
npm run test-connection
Note: If you get a "'tsx' is not recognized" error, the test script will automatically build the project first and use Node.js instead.
Usage
Development
npm run dev
Production
npm run build
npm start
Adding to Claude Desktop
-
Find your Claude Desktop config file:
- Windows:
%APPDATA%\Claude\claude_desktop_config.json - macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Linux:
~/.config/Claude/claude_desktop_config.json
- Windows:
-
Add the MCP server configuration:
Quick setup (Recommended)
{ "mcpServers": { "questrade": { "command": "npx", "args": ["questrade-mcp-server"], "env": { "QUESTRADE_REFRESH_TOKEN": "your_refresh_token_here" } } } }Local development build
{ "mcpServers": { "questrade": { "command": "node", "args": ["/path/to/your/project/dist/index.js"], "env": { "QUESTRADE_REFRESH_TOKEN": "your_refresh_token_here" } } } } -
If using local build, update the path to match your actual project location
-
Restart Claude Desktop
-
Test the connection by asking Claude to show your Questrade accounts
For detailed setup instructions, see claude-desktop-config.md.
Available Tools
Account Management
get_accounts- Get all Questrade accountsget_positions- Get positions for a specific accountget_balances- Get balances for a specific accountget_orders- Get order history for an account
Market Data
search_symbols- Search for symbols by prefixget_symbol- Get detailed symbol informationget_quotes- Get real-time quotes for symbolsget_candles- Get historical price data
Authentication
refresh_token- Refresh the API access token
Built-in Prompts
The MCP server includes helpful prompts for common trading analysis tasks:
Portfolio Summary
Prompt: portfolio_summary
- Get a comprehensive portfolio analysis with account balances, positions, and performance
- Optional: Specify
accountNumber(uses first account if not provided)
Stock Analysis
Prompt: stock_analysis
- Analyze a specific stock with current quotes, symbol information, and recent performance
- Required:
symbol(e.g., "AAPL", "TSLA", "MSFT")
Trading Opportunities
Prompt: trading_opportunities
- Identify potential trading opportunities based on current positions and market data
- Optional:
accountNumber(uses first account if not provided) - Optional:
riskLevel("conservative", "moderate", or "aggressive")
Example Usage
Simply ask Claude:
- "Use the portfolio_summary prompt to analyze my trading account"
- "Analyze AAPL stock using the stock_analysis prompt"
- "Show me trading opportunities with conservative risk level"
Tool Examples
Get Accounts
{
"name": "get_accounts"
}
Get Positions
{
"name": "get_positions",
"arguments": {
"accountNumber": "12345678"
}
}
Search Symbols
{
"name": "search_symbols",
"arguments": {
"prefix": "AAPL",
"offset": 0
}
}
Get Quotes
{
"name": "get_quotes",
"arguments": {
"symbolIds": [8049, 9291]
}
}
Configuration
The server uses environment variables for configuration:
QUESTRADE_API_URL: Base URL for Questrade API (default: https://api01.iq.questrade.com)QUESTRADE_REFRESH_TOKEN: Your API refresh tokenQUESTRADE_TOKEN_DIR: Custom directory for token storage (default:~/.questrade-mcp)
Error Handling
The server includes comprehensive error handling for:
- Invalid or expired tokens (automatic refresh)
- Missing required parameters
- API rate limits and network errors
- Invalid account numbers or symbol IDs
Security Notes
- Never commit your
.envfile to version control - Access tokens expire after 7 days
- Refresh tokens are automatically used to get new access tokens
- This is an unofficial tool - ensure you comply with Questrade's API terms of service
- Always verify trading decisions independently before executing trades
Development
Project Structure
src/
├── index.ts # Main MCP server implementation
├── questrade-client.ts # Questrade API client
└── types.ts # TypeScript type definitions
Building
npm run build
Cleaning
npm run clean
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。