marketcalls/openalgo

Open Source Algo Trading Platform for Everyone

What it solves

OpenAlgo provides a unified, self-hosted trading platform that eliminates the need to rewrite strategy code when switching between different brokers. It solves the complexity of hosting and executing algorithmic trading strategies by providing a built-in environment for both coders and non-coders, while maintaining complete data privacy.

How it works

The platform acts as a comprehensive trading surface that integrates with 36 different broker plugins. It uses a unified REST API and a ZeroMQ-based WebSocket proxy for normalized market data and order management. Users can interact with the platform through several interfaces:

  • Python Strategy Host: A browser-based editor for running and scheduling Python scripts with process isolation.
  • Flow: A no-code, node-based visual strategy builder.
  • AI Agent: An LLM-powered chat interface that can analyze market data, draw charts, and suggest trades (requiring explicit user approval).
  • Options Suite: A set of twelve analytical tools for calculating Greeks, IV smiles, and payoff diagrams.

Who it’s for

  • Algorithmic Traders: Those who want to host and run Python strategies without managing separate servers or cron jobs.
  • No-Code Traders: Individuals who prefer a visual drag-and-drop interface to build trading logic.
  • Options Traders: Traders needing advanced analytics like Gamma Exposure (GEX) and Volatility Surfaces.
  • Developers: Those building trading apps who need a single API contract to support multiple brokers.

Highlights

  • Unified Broker API: A single contract supporting 36 different securities and crypto derivatives brokers.
  • AI Integration: Built-in LLM agent and MCP server for natural language trading via Claude, ChatGPT, or local models via Ollama.
  • Sandbox Mode: An "API Analyzer" with ₹1 Crore virtual capital for risk-free strategy testing.
  • Self-Hosted: Complete privacy with zero data collection, utilizing a local SQLite and DuckDB stack.
  • Multi-Platform Support: Direct integrations with TradingView, Amibroker, ChartInk, and various official SDKs.

Related

  • Project
    runkids/skillshare

    A centralized management tool that synchronizes AI CLI skills, agents, and rules across multiple AI tools from a single source of truth.

  • Project
    rebelytics/one-skill-to-rule-them-all

    A Claude‑skill that watches your work sessions, logs observations, and suggests new or improved Claude skills—including improvements to itself. Install by adding the `.skill` bundle to Claude, let it run, then review its structured logs to keep your skill library evolving automatically.

  • Project
    google-deepmind/mctx

    A JAX-native library providing high-performance implementations of Monte Carlo tree search algorithms like AlphaZero and MuZero for reinforcement learning.

  • Project
    ml-explore/mlx-swift-examples

    A collection of example programs and libraries for MLX Swift, demonstrating how to build LLM, VLM, and image generation apps on iOS and macOS.