yzfly/Awesome-AGI-Agents

🤖 Awesome list of AGI Agents. Agents 精选资源合集.

What it solves

Keeping up with the fast-moving AI agent ecosystem is hard. This project gathers a wide range of resources — articles, videos, research papers, significant projects, development platforms, and curated lists — into one place. It helps developers, researchers, and enthusiasts quickly find high-quality material about building, evaluating, and deploying AI agents.

How it works

The repository is organized into clear sections. It starts with a table of contents, then lists articles and videos that explain agent concepts and industry trends. Next come academic papers and surveys about agent architecture and benchmarks. A major section showcases important projects from 2024 to 2026, including frameworks like LangGraph, CrewAI, and OpenAI Agents SDK, plus protocols like MCP and A2A. There is also a "frontier projects" section for experimental agents, and a section dedicated to agent development platforms. Finally, it points to other curated lists for deeper exploration.

Who it's for

Anyone interested in AI agents — from beginners looking for introductory articles to experienced developers seeking production-ready frameworks. Researchers can find relevant papers and benchmarks, while hobbyists can explore cutting-edge projects. The content is especially useful for those who want to track the evolution of the agent ecosystem over time.

Highlights

  • Covers both foundational knowledge (articles, papers) and practical tools (frameworks, platforms).
  • Includes notable 2024-2026 projects like OpenAI Agents SDK, Google ADK, and Microsoft Agent Framework.
  • Documents key industry protocols such as Model Context Protocol (MCP) and Agent2Agent (A2A).
  • Features a curated list of other awesome lists, helping users discover even more specialized resources.
  • Continuously updated, with notes on recent additions and star history tracking.

Related

  • Project
    SJTU-IPADS/SkVM

    SkVM is a compilation and runtime system that makes LLM agent skills portable and optimized across different models and agent harnesses.

  • Project
    oughtinc/ice

    ICE is a Python library for building, running, and visualising structured language‑model workflows (recipes) with interchangeable agents and a browser‑based trace UI.

  • Project
    google-agentic-commerce/a2a-x402

    The A2A x402 Extension brings cryptocurrency payments to the Agent-to-Agent (A2A) protocol, letting agents monetize their services through on-chain transactions. It defines a standardized payment flow so any A2A agent can charge for API calls, data processing, AI inference, or other capabilities.

  • Project
    joinly-ai/joinly

    A connector middleware that enables AI agents to join and participate in video calls on platforms like Zoom, Google Meet, and Microsoft Teams via voice and chat.