turtlestack-lite
TurtleStack Trading MCP Server is a comprehensive Model Context Protocol (MCP) server that provides unified access to multiple Indian stock brokers through Claude AI. It enables seamless trading operations, portfolio management, and advanced technical analysis across Kite (Zerodha), Groww, Dhan, and AngelOne brokers.
claude mcp add --transport stdio turtlehq-tech-turtlestack-lite node /path/to/turtlestack-lite/src/index.js
How to use
TurtleStack Lite is a multi-broker trading MCP server that provides a unified API to interact with Kite (Zerodha), Groww, and Dhan across real-time trading, portfolio management, and technical indicators. It exposes a modular broker architecture where each broker is implemented separately and can be extended with additional brokers. The server includes a suite of technical indicators (40+), supports Claude AI integration for natural language trading commands, and can be deployed with Cloudflare Workers for edge computing. To use it, start the MCP server and connect your Claude AI configuration or any client that understands the MCP protocol to perform actions such as placing orders, querying positions, and retrieving market data across multiple brokers from a single endpoint.
How to install
Prerequisites:
- Node.js 18+ installed on your system
- Git installed
- Access to the project repository (git)
Install and run locally:
# 1. Clone the repository
git clone https://github.com/turtlehq-tech/turtlestack-lite.git
cd turtlestack-lite
# 2. Install dependencies
npm install
# 3. Start the MCP server
npm start
Alternative development workflow:
# 1. Install dependencies
npm install
# 2. Run in development mode (if defined)
npm run dev
Configuration notes:
- Update any placeholder credentials before running in a real environment.
- If using Claude AI integration, ensure your Claude configuration points to the MCP server (see Claude AI configuration example in the repo).
Additional notes
Tips and common considerations:
- The MCP config example uses a Node-based server entry point at src/index.js. Replace the path with your actual installation path when integrating with Claude or other clients.
- This project uses dummy placeholders for sensitive credentials; replace them with your real credentials before usage.
- For cloud deployments (Cloudflare Workers), follow the Cloudflare-specific configuration guidance in the README to set up workers and KV namespaces.
- Security: credentials should be provided at runtime through Claude commands or environment variables rather than stored in configuration files.
- If you encounter port or network issues, ensure your firewall allows the MCP server to communicate with broker APIs and Claude.
Related MCP Servers
mcp-aktools
📈 提供股票、加密货币的数据查询和分析功能MCP服务器
anytype
An MCP server enabling AI assistants to interact with Anytype - your encrypted, local and collaborative wiki - to organize objects, lists, and more through natural language.
flapi
API Framework heavily relying on the power of DuckDB and DuckDB extensions. Ready to build performant and cost-efficient APIs on top of BigQuery or Snowflake for AI Agents and Data Apps
storyblok
A modular, extensible MCP Server for managing Storyblok spaces, stories, components, assets, workflows, and more via the Model Context Protocol (MCP).
openproject
A Model Context Protocol (MCP) server that provides seamless integration with OpenProject API v3.
zerodha
Zerodha MCP Server & Client - AI Agent (w/Agno & w/Google ADK)