Questrade MCP Server
Enables querying portfolio data, positions, balances, quotes, price history, orders, and account activity through the Questrade API.
README
Questrade MCP Server
An MCP server for the Questrade trading API. Use it from MCP-compatible clients (Claude, ChatGPT, Cursor, and others) to inspect portfolio data, quotes, market history, orders, and account activity.
This server is designed around outcome-oriented tools: each tool performs the necessary Questrade API orchestration internally so agents can get complete results with fewer round-trips.
Table of Contents
- Quick Start
- MCP Features
- Tools
- Configuration
- Token Management
- Real-Time Data Notes
- Development Setup
- Project Structure
- Docker
- CI/CD
- Security
- Contributing
- Security Policy
- Release Process
- License
Quick Start
Prerequisites
- Node.js 22+
- A Questrade account with API access enabled
1) Clone, Install, Build
git clone https://github.com/DrWheelicus/questrade-mcp-server.git
cd questrade-mcp-server
npm install
npm run build
2) Get a Questrade Refresh Token
- Open the Questrade API Hub
- Register a personal app (or reuse an existing one)
- Enter a name for the device and a description
- Click Create
- Select New device
- Select Generate new token
- Copy the refresh token and paste it into the .env file
[!WARNING] The refresh token is a secret and should not be shared. If you lose it, you will need to generate a new one. (Steps 5-7)
3) Add Server to Your MCP Client
Use a config like this (replace with your local absolute path):
{
"mcpServers": {
"questrade": {
"command": "node",
"args": ["/absolute/path/to/Questrade/dist/index.js"],
"env": {
"QUESTRADE_REFRESH_TOKEN": "<your-token>"
}
}
}
}
Common locations:
- Claude Desktop:
~/Library/Application Support/Claude/claude_desktop_config.json(macOS) or%APPDATA%\Claude\claude_desktop_config.json(Windows) - Claude Code: project
.mcp.jsonor~/.claude/mcp.json - Cursor: Settings -> MCP Servers
[!NOTE] On some Windows MSIX installs, Claude Desktop may read a virtualized config path instead:
%LOCALAPPDATA%\Packages\Claude_pzs8sxrjxfjjc\LocalCache\Roaming\Claude\claude_desktop_config.json.
4) Verify
Restart your MCP client and ask:
Show me my portfolio
MCP Features
- Tools: purpose-built portfolio and market data operations
- Transport support:
stdiofor local clients, Streamable HTTP for remote clients - Validation: tool inputs validated with Zod schemas
- Token lifecycle management: encrypted persistence and automatic refresh
This server currently exposes MCP tools (not prompts/resources).
Tools
| Tool | Description |
|---|---|
getPortfolio |
Full portfolio overview across accounts with positions and balances |
getPositions |
Positions held in a specific account |
getBalances |
Cash balances and buying power for an account |
lookupSymbol |
Search by ticker/name and return symbol details plus live quote |
getQuotes |
Batch Level 1 quotes by ticker names or symbol IDs |
getPriceHistory |
Historical OHLCV candles at intervals from 1 minute to yearly |
getOrders |
Order history with state and date filters |
getAccountActivity |
Combined activities and executions timeline |
Configuration
All runtime configuration is environment-driven:
| Variable | Required | Default | Description |
|---|---|---|---|
QUESTRADE_REFRESH_TOKEN |
Yes (first run) | - | OAuth refresh token from Questrade |
QUESTRADE_ENVIRONMENT |
No | production |
production or practice |
QUESTRADE_MCP_TRANSPORT |
No | stdio |
stdio or http |
QUESTRADE_MCP_PORT |
No | 3100 |
Port for HTTP transport |
QUESTRADE_ENCRYPTION_KEY |
No | machine-derived | Override encryption key for token storage |
LOG_LEVEL |
No | info |
fatal, error, warn, info, debug, trace |
The CLI flag --transport overrides QUESTRADE_MCP_TRANSPORT.
HTTP Mode (Remote Clients)
QUESTRADE_REFRESH_TOKEN=<your-token> node dist/index.js --transport=http
Server endpoint:
http://localhost:3100/mcp
Token Management
Questrade uses single-use refresh tokens. This server:
- Exchanges the initial refresh token on first startup
- Encrypts and stores token material with AES-256-GCM
- Renews access proactively before expiry
- Retries once on
401responses after refreshing
Token storage paths:
| Platform | Path |
|---|---|
| Windows | %APPDATA%\questrade-mcp\token.enc |
| macOS | ~/Library/Application Support/questrade-mcp/token.enc |
| Linux | ~/.config/questrade-mcp/token.enc |
After first successful startup, QUESTRADE_REFRESH_TOKEN is only needed again if the cached token expires.
Real-Time Data Notes
Questrade requires a paid market data subscription for real-time quotes.
Without one, getQuotes returns snap quotes with daily exchange limits and may
fall back to delayed data. Use isDelayed in quote responses to detect this.
Development Setup
npm install
npm run dev
npm run build
npm run lint
npm run typecheck
npm test
MCP Inspector
npx @modelcontextprotocol/inspector node dist/index.js
Project Structure
.
├── src/
│ ├── index.ts # Server entrypoint and transport selection
│ ├── server.ts # MCP server and tool registration
│ ├── config.ts # Env/config parsing
│ ├── log.ts # Logger setup
│ ├── auth/ # Token persistence and refresh
│ ├── client/ # Questrade API client and rate limiting
│ ├── tools/ # MCP tool implementations
│ ├── transports/ # stdio and HTTP transport handlers
│ └── types/ # Shared API/domain types
├── tests/ # Unit/integration tests
├── .github/workflows/ # CI/CD workflows
└── Dockerfile
Docker
Build and run HTTP transport:
docker build -t questrade-mcp .
docker run -p 3100:3100 -e QUESTRADE_REFRESH_TOKEN=<your-token> questrade-mcp
CI/CD
- CI on push/PR to
main: type check, lint, build, tests,npm audit, dependency review - CD on tags matching
v*.*.*: multi-arch Docker image to GHCR and GitHub Release
Security
- Token material encrypted at rest (AES-256-GCM)
- Refresh tokens rotated by design (Questrade behavior)
- Token values redacted from logs
- Read-only server scope (no trade placement tools)
- Input validation on all tool parameters (Zod)
- HTTP transport uses per-client session isolation
- Docker image runs as non-root user
Contributing
See CONTRIBUTING.md for contributor setup, quality checks, and PR process.
GitHub issue and PR templates are provided in .github/ to keep reports and reviews consistent.
Security Policy
See SECURITY.md for how to report vulnerabilities privately.
Release Process
See RELEASE.md for the tag-based release checklist and verification steps.
License
Proprietary - All Rights Reserved.
No permission is granted to copy, modify, redistribute, sublicense, or use this code except with explicit written permission from the copyright holder.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.