MyAgent

What is MyAgent

A from-scratch general-purpose Agent Runtime supporting Agent Loop, permission control, MCP, sub-Agents, context folding and self-evolving Skills.

Key features

• Implements a general-purpose Agent Runtime that unifies model inference orchestration, tool calling, permission control, long-term memory, Skills and context management
• Supports a complete Agent Loop including model requests, tool call parsing, permission checks, tool execution, result writing back and session saving
• Provides a tool system with permission control, supporting read/write files, precise editing, search, Shell, Skill calling, sub-Agents and MCP tools
• Implements a Skills system that stores reusable methods via SKILL.md, supporting retrieval, calling, versioned evolution and self-evolution
• Supports MCP external tool extensions, sub-Agents (explore, plan, general etc.), session restoration and context compression

Who it is for

Suitable for developers learning the underlying mechanisms of general Agents, and as a foundation for secondary development of personal Agents, project analysis assistants or domain-specific Agents.

How to get started

1. Clone the GitHub repository ztt0216/MyAgent
2. Run python3 -m venv .venv and activate the virtual environment
3. Execute pip install -r requirements.txt to install dependencies
4. Configure the .env file and start the CLI or REPL via agents/main.py

Compiled from public descriptions on the tool's official website / README / launch page. Pricing and user numbers have not been verified; please check the official website. Some sections were drafted with AI assistance. Written in Chinese and English; other languages are machine-translated.