一个从零实现 AI Agent 的开源学习项目——每一步都有源码，每一章都能运行，渐进式理解智能体。Build an AI Agent from Scratch — Learn AI Agents step by step with runnable source code.

agents ai-agent llm-agents agentic-ai python-agents agent-tutorial agent-from-scratch python-agentic-ai

• Updated Aug 6, 2026
• JavaScript

CXS is a deterministic continuity standard for long-range, multi-session, and long-chat LLM reasoning. It enables GPT, Claude, Grok, Gemini, and local models to share state through a human-readable Continuity Passport with strict-mode rules, drift detection/repair, and full chain-of-custody. Validated across 250+ continuity tests.

machine-learning ai deep-learning drift drift-detection large-language-models llm llm-training reasoning-engine memory-architecture ai-continuity continuity-theory python-agents ai-continuity-prompt-engineering continuity-ai continuity-architecture

• Updated Dec 20, 2025
• TeX

Local-first LLM operating-system hub for BACH, Rinnsal, gardener, SQLite agent memory, MCP servers, skills, and multi-agent orchestration

open-source automation sqlite mcp multi-agent ai-agents local-first agent-framework llm local-ai llm-agents llm-memory llm-os text-based-os model-context-protocol mcp-server claude-code python-agents agent-os sqlite-memory

• Updated Sep 20, 2026

Python toolkit for parallel Claude and LLM agent orchestration with local state: consensus voting, stigmergy, and boss-worker workflows

python consensus orchestration multi-agent parallel-processing multi-agent-systems swarm-intelligence ai-agents stigmergy local-first agent-framework llm anthropic llm-agents llm-orchestration claude-code claude-agents python-agents parallel-llm consensus-voting

• Updated Nov 3, 2026
• Python

This repository showcases a variety of AI agent implementations, ranging from chat applications and RAG (Retrieval-Augmented Generation) pipelines to specialized MCP (Model Context Protocol) servers.

ai mcp ai-agents rag rag-agents autonomus-agents google-adk python-agents

• Updated Dec 2, 2025
• Jupyter Notebook

an example project demonstrating how to build, define, and integrate custom skills with the Google ADK (Agent Development Kit) 2.0 framework. The repository features a step-by-step implementation of a Python-based agent (skills_app/agent.py) that programmatically invokes external skill scripts (such as a lucky color lookup) to handle user prompts.

skills python-agent agent-development-kit google-adk python-agents adk-agent-2-0

• Updated Jun 4, 2026
• Python

Top-tier contribution fork for smolagents: secure coded-agent patterns, MCP interoperability, and maintainer-friendly contributor guidance.

open-source security agentic-ai smolagents python-agents

• Updated Jun 19, 2026
• Python

From zero-shot prompts to multi-agent loops — the complete progression of AI engineering in 2026. 7 levels, runnable Python code, real patterns used in production systems.

tutorial multi-agent correction ai-agents llm prompt-engineering chain-of-thought react-agent python-agents loop-agents loop-engineering llm-tutorial-2026

• Updated Jul 17, 2026
• Python