Agent Systems Handbook（智能体系统手册）

import SupportCTA from "/snippets/support-cta.mdx";

Best for people who want a high-level understanding of agent systems without committing to a strict build sequence.

Recommended reads

• The Agent Systems: start

  with the core definition and operating loop.

• History Of Agent Ideas: use the short historical arc to understand why the field looks the way it does.

• Agents Vs Workflows: clarify the boundary between deterministic orchestration and bounded autonomy.

• Deep Research Agents: see how one of the clearest modern agent products is structured.

• Ecosystem: compare the major tools, platforms, and market categories.

• Radar: track what is changing quickly across the field.

Optional deeper dives

• Agent Memory And Retrieval: go here if you want a practical mental model for memory, retrieval, and artifacts.
• Reasoning And Control Patterns: go here if you want the control-loop view of agent behavior.
• LLM Foundations For Agent Systems: go here if you want the minimum model literacy behind modern agents.
• Context Engineering: go here if you want the production-minded view of how context is managed over time.
• AI Education Policy To Practice: go here if you want a product-agnostic example of turning AI policy language into a practical learning activity.

What this guide optimizes for

• Clear orientation without requiring a fixed sequence
• High-level understanding before implementation detail
• A broad view of the field and its moving parts