TastyScanner MCP Server
Enables AI agents to scan markets, build option strategies, track positions, and execute trades on TastyTrade via the Model Context Protocol.
README
<div align="center">
šÆ TastyScanner MCP Server
Let Claude trade options for you on TastyTrade
Features ⢠Quick Start ⢠Documentation ⢠Discord ⢠Sponsor
</div>
š¤ What is this?
TastyScanner MCP Server connects Claude, GPT, or any AI agent to your TastyTrade account via the Model Context Protocol.
Ask your AI assistant to:
- "Scan the market for high IVR stocks"
- "Build an Iron Condor on SPY with 30-45 DTE"
- "Show my positions and P&L"
- "Close my AAPL position at 50% profit"
<!--
TODO: Add demo GIF here
-->
⨠Features
| Feature | Description |
|---|---|
| š Market Scanner | Scan symbols for IVR, IV, beta, price, earnings dates |
| šļø Strategy Builder | Auto-construct Iron Condors with POP, R:R, Greeks |
| š Position Tracking | Real-time P&L, Greeks, days to expiration |
| ā” Trade Execution | Place and close multi-leg options orders |
| š Safety First | Live trading disabled by default, sandbox support |
| š³ Docker Ready | One-command deployment |
šÆ Why TastyScanner?
| TastyScanner | Other MCP Servers | |
|---|---|---|
| Strategy Builder (Iron Condor auto-construction) | ā | ā |
| POP Calculation (Probability of Profit) | ā | ā |
| TypeScript (Node.js ecosystem) | ā | Python only |
| Docker-first | ā | Manual setup |
| Watchlist Management | ā | ā |
š Quick Start
Option 1: Docker (Recommended)
# Clone
git clone https://github.com/technet365/TastyScanner-MCP-Server.git
cd TastyScanner-MCP-Server
# Configure
cp .env.example .env
# Edit .env with your TastyTrade OAuth credentials
# Run
docker compose up -d
# Verify
curl http://localhost:7698/health
Option 2: Node.js
npm install
cp .env.example .env
npm run build
npm start
Connect to Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"tastytrade": {
"url": "http://localhost:7698/mcp"
}
}
}
Architecture
āāāāāāāāāāāāāāāāāāāāāā āāāāāāāāāāāāāāāāāāāāāāā
ā Claude / GPT / ā ā TastyScanner-MCP ā
ā Any AI Agent āāā MCP āāā (port 7698) ā
āāāāāāāāāāāāāāāāāāāāāā HTTP āāāāāāāāāāāā¬āāāāāāāāāāā
ā
@tastytrade/api
ā
āāāāāāāāāāāā¼āāāāāāāāāāā
ā TastyTrade API ā
ā (WebSocket + REST) ā
āāāāāāāāāāāāāāāāāāāāāāā
The MCP server connects to TastyTrade using the official @tastytrade/api SDK
with OAuth authentication (client credentials + refresh token).
Quick Start
1. Configure credentials
cp .env.example .env
# Edit .env with your TastyTrade OAuth credentials
2. Build and run
docker compose build
docker compose up -d
curl http://localhost:7698/health
3. Connect to Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"tastytrade": {
"url": "http://localhost:7698/mcp"
}
}
}
Available Tools
| Tool | Description |
|---|---|
get_market_overview |
Scan symbols for IVR, IV, beta, price, earnings |
get_strategies |
Build Iron Condor setups for a symbol |
get_positions |
List current open positions with P&L |
execute_trade |
Place an options order (ā ļø real money) |
close_position |
Close an existing position |
adjust_order |
Adjust working order price for better fill |
get_working_orders |
List pending/unfilled orders |
get_account_info |
Account balance and buying power |
get_connection_status |
Check TastyTrade connection health |
get_watchlists |
List personal and platform watchlists |
manage_watchlist |
Create, add to, remove from, or delete watchlists |
Tool Details
get_market_overview
Params: symbols?[], min_ivr?, min_price?, max_price?
Returns: [{symbol, price, ivr, iv, beta, earnings_date}] sorted by IVR desc
get_strategies
Params: symbol, min_dte?, max_dte?, min_delta?, max_delta?, wings?[], max_results?
Returns: [{strategy_name, expiry_date, dte, legs[], credit, max_profit, max_loss, rr_ratio, pop, theta, delta, wings, bpe}]
get_positions
Params: (none)
Returns: [{position_id, symbol, strategy, legs[], entry_credit, current_value, pnl, pnl_percent, dte, opened_at}]
execute_trade
Params: symbol, legs[{action, symbol, quantity}], limit_price, price_effect, time_in_force?, order_type?
Returns: {order_id, status, message}
ā ļø This places REAL orders with REAL money.
close_position
Params: position_id, reason, limit_price?
Returns: {order_id, status, pnl_realized, message}
adjust_order
Params: order_id, adjustment ('improve_fill' | 'custom'), custom_price?
Returns: {order_id, old_price, new_price, status, message}
ā ļø Requires ENABLE_LIVE_TRADING=true
get_working_orders
Params: (none)
Returns: [{order_id, symbol, status, price, price_effect, legs[]}]
get_watchlists
Params: include_public? (default: true)
Returns: {personal: [{name, symbols[]}], platform: [{name, symbol_count}]}
manage_watchlist
Params: action ('create' | 'add' | 'remove' | 'delete'), name, symbols?[]
Returns: {success, message}
Development
Run locally (without Docker)
cd tastyscanner-mcp
npm install
# Copy env template and fill in credentials
cp .env.example .env
# Edit .env with your OAuth credentials
# Run with hot-reload
npm run dev
# Or build and run
npm run build
npm start
Test with curl
# Initialize MCP session (add -H "Authorization: Bearer <token>" if MCP_AUTH_TOKEN is set)
curl -X POST http://localhost:7698/mcp \
-H "Content-Type: application/json" \
-H "Authorization: Bearer your-auth-token" \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "initialize",
"params": {
"protocolVersion": "2025-03-26",
"capabilities": {},
"clientInfo": {"name": "test", "version": "1.0.0"}
}
}'
# Call a tool (use session-id from init response)
curl -X POST http://localhost:7698/mcp \
-H "Content-Type: application/json" \
-H "Authorization: Bearer your-auth-token" \
-H "mcp-session-id: <session-id>" \
-d '{
"jsonrpc": "2.0",
"id": 2,
"method": "tools/call",
"params": {
"name": "get_connection_status",
"arguments": {}
}
}'
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
TASTY_CLIENT_ID |
Yes | ā | TastyTrade OAuth Client ID |
TASTY_CLIENT_SECRET |
Yes | ā | TastyTrade OAuth Client Secret |
TASTY_REFRESH_TOKEN |
Yes | ā | TastyTrade OAuth Refresh Token |
TASTY_ACCOUNT |
No | auto | Account number (auto-detects first) |
TASTY_PRODUCTION |
No | true |
true = production, false = sandbox |
MCP_PORT |
No | 7698 |
HTTP server port |
MCP_AUTH_TOKEN |
No | ā | Bearer token for endpoint auth. If set, all /mcp requests require Authorization: Bearer <token> |
LOG_LEVEL |
No | info |
debug, info, warn, error |
ENABLE_LIVE_TRADING |
No | false |
Set true to allow execute_trade and adjust_order |
Security Notes
- Authentication: Set
MCP_AUTH_TOKENin.envto require Bearer token auth on all/mcpendpoints. Without it, the server is open (backwards compatible but not recommended for production). - Rate limiting: 120 requests/minute per IP on MCP endpoints.
- CORS: Restricted to
MCP_CORS_ORIGIN(defaulthttp://localhost:3333). - Credentials are passed via environment variables, never hardcoded.
execute_tradeandadjust_orderrequireENABLE_LIVE_TRADING=trueā disabled by default.- Session IDs use
crypto.randomUUID()(cryptographically secure). - Account numbers are masked in logs (last 4 digits only).
- Do not expose port 7698 to the internet without auth enabled.
š Support the Project
If TastyScanner MCP Server saves you time or helps you trade better, consider supporting its development:
<a href="https://github.com/sponsors/technet365"> <img src="https://img.shields.io/badge/Sponsor-ā¤ļø-red?style=for-the-badge&logo=github" alt="Sponsor on GitHub"> </a> <a href="https://discord.gg/739SVUQBrE"> <img src="https://img.shields.io/badge/Discord-Join-7289da?style=for-the-badge&logo=discord&logoColor=white" alt="Join Discord"> </a>
Sponsor Benefits
| Tier | Benefits |
|---|---|
| ā $5/mo | Support development, name in README |
| š„ $15/mo | Early access to new features, priority support |
| š„ $50/mo | Direct input on roadmap, 1:1 onboarding call |
š License
MIT Ā© 2024-2025 technet365
<div align="center">
Made with ā¤ļø for the TastyTrade community
</div>
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.