Welcome to Open Agent Tools - a collection of open-source tools and frameworks for building AI agents, trading systems, and developer utilities.

• basic-open-agent-tools - An open foundational toolkit providing essential components for building AI agents with minimal dependencies for local (non-HTTP/API) actions. Designed with agent-friendly type signatures to eliminate "signature too complex" errors, while offering core utilities that developers can easily integrate into their agents to avoid excess boilerplate.

  • Language: Python
  • License: MIT

• any-agent - Flexible Python framework for building AI agents compatible with multiple LLM providers. Enables agent development with provider-agnostic implementations for maximum portability.

  • Language: Python
  • License: MIT

• Basic-Agent-Chat-Loop - A feature-rich, interactive CLI for AI agents with token tracking, prompt templates, agent aliases, command history, and multi-line input support. Provides a flexible command-line chat framework for building AI agents with advanced interaction capabilities.

  • Language: Python
  • License: MIT

• open-stocks-mcp - An MCP (Model Context Protocol) server providing access to stock market data and trading capabilities through Robin Stocks API.

  • Language: Python
  • License: Apache-2.0

• open-paper-trading-mcp - A comprehensive paper trading simulator with dual interfaces: REST API (FastAPI) and AI agent tools (MCP). Designed for algorithmic trading development, strategy backtesting, options trading simulation, and training AI agents in realistic market environments without financial risk.

  • Language: Python
  • License: Apache-2.0

• coding-open-agent-tools - Advanced code generation and shell scripting toolkit for AI agents. Provides parsing, validation, and analysis tools that reduce token consumption by handling deterministic operations. Includes 154+ functions across Git operations, database tools, Python code analysis, shell validation, performance profiling, and quality analysis.

  • Language: Python
  • License: MIT

• claude_slash_commands_library - A collection of powerful slash commands for Claude Code to enhance your development workflow.

  • License: Apache-2.0

• agent-examples - Example implementations and demonstrations of AI agents using Google ADK, Strands, LangChain, and other frameworks.

  • Language: Python
  • License: MIT

We build open-source tools that make AI agent development accessible, reliable, and powerful. Our focus areas include:

• Agent Frameworks - Core libraries for building AI agents
• Trading Systems - Tools for financial data analysis and paper trading
• Developer Utilities - Extensions and libraries that enhance the development experience

We welcome contributions to any of our projects! Each repository has its own contribution guidelines and setup instructions.

Our projects use a mix of MIT and Apache-2.0 licenses. Check individual repositories for specific licensing details.