Agent_In_Action and all-agentic-architectures

Project A, a practical development handbook for agentic AI, would likely use the implementations from Project B, an extensive collection of agentic architectures, making them complements.

Agent_In_Action
53
Established
all-agentic-architectures
51
Established
Maintenance 10/25
Adoption 9/25
Maturity 13/25
Community 21/25
Maintenance 2/25
Adoption 10/25
Maturity 15/25
Community 24/25
Stars: 77
Forks: 33
Downloads:
Commits (30d): 0
Language: Jupyter Notebook
License: MIT
Stars: 2,892
Forks: 517
Downloads:
Commits (30d): 0
Language: Jupyter Notebook
License: MIT
No Package No Dependents
Stale 6m No Package No Dependents

About Agent_In_Action

FlyAIBox/Agent_In_Action

Agentic AI 智能体开发实战

Implements a complete agentic AI system using LangGraph and MCP protocols, covering architecture (GAME framework with Goals/Actions/Memory/Environment), tool integration patterns, and production deployment via FastAPI + Streamlit with Docker. Includes six end-to-end projects spanning multi-agent orchestration, monitoring via Langfuse, and domain-specific model fine-tuning with LlamaFactory and vLLM inference optimization.

About all-agentic-architectures

FareedKhan-dev/all-agentic-architectures

Implementation of 17+ agentic architectures designed for practical use across different stages of AI system development.

Implements 17+ architectures using LangGraph for stateful agent orchestration, with each pattern (Reflection, ReAct, Planning, Multi-Agent, Tree of Thoughts, etc.) demonstrated end-to-end in executable Jupyter notebooks. Features integrated evaluation via LLM-as-a-Judge for quantitative performance measurement and real-world scenarios spanning financial analysis, coding, and medical triage. Designed as a structured learning path progressing from single-agent enhancements through advanced multi-agent collaboration, memory systems, and self-aware agents with built-in safety mechanisms.

Scores updated daily from GitHub, PyPI, and npm data. How scores work