Mini-Agent is a minimal yet professional demo project by MiniMax that showcases best practices for building AI agents using the MiniMax M2.5 model.

🔗 GitHub: https://github.com/MiniMax-AI/Mini-Agent

Key Features

  • Full Agent Execution Loop — Complete foundation with file system and shell operation toolset
  • Persistent Memory — Session Note Tool retains key information across multiple sessions
  • Intelligent Context Management — Auto-summarizes conversation history to handle contexts up to configurable token limits, enabling infinitely long tasks
  • Claude Skills Integration — Comes with 15 professional skills for documents, design, testing, and development
  • MCP Tool Integration — Natively supports MCP for knowledge graph access and web search
  • Comprehensive Logging — Detailed logs for every request, response, and tool execution
  • Clean & Simple Design — Beautiful CLI with an easy-to-understand codebase

Why It Matters

Mini-Agent leverages an Anthropic-compatible API with full support for interleaved thinking, unlocking M2.5 powerful reasoning capabilities for long, complex tasks. It supports both global and China platforms, making it accessible worldwide.

Quick Start

# Install directly from GitHub
uv tool install git+https://github.com/MiniMax-AI/Mini-Agent.git

# Run setup
curl -fsSL https://raw.githubusercontent.com/MiniMax-AI/Mini-Agent/main/scripts/setup-config.sh | bash

# Configure your API key and start
mini-agent

Great project for anyone looking to build production-grade AI agents with a clean, minimal codebase. Worth checking out! ⭐