# Agent 架构与工程

## ToolCall
> [Toolformer: Language Models Can Teach Themselves to Use Tools](https://arxiv.org/abs/2302.04761)

## Agent 架构
> [Building Effective AI Agents \ Anthropic](https://www.anthropic.com/engineering/building-effective-agents)

### ReAct
> [ReAct: Synergizing Reasoning and Acting in Language Models](https://arxiv.org/abs/2210.03629)


### Plan-and-Execute
> [Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models](https://arxiv.org/abs/2305.04091)


### Reflection
> [Reflexion: Language Agents with Verbal Reinforcement Learning](https://arxiv.org/abs/2303.11366)

### MultiAgent
> [AutoGen](https://arxiv.org/abs/2308.08155), [MetaGPT](https://arxiv.org/abs/2308.00352)


## Context Engineer
> [Context Engineering for Agents](https://rlancemartin.github.io/2025/06/23/context_engineering/)

## Harness Engineer
> [harness-engineering-OpenAI](https://openai.com/zh-Hans-CN/index/harness-engineering/)

## Loop Engineer

## 参考阅读

- [从ToolCall到Harness、Claw-bilibili](https://www.bilibili.com/video/BV1dw526tEMA)
- [Agent 架构 - 菜鸟教程](https://www.runoob.com/ai-agent/agent-architecture.html)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://arxiv.org/abs/2201.11903)

**claude code**
- [learn claude](https://claude.nagdy.me/learn/)
- [Claude Code Unpacked](https://ccunpacked.dev/)
- [深入学习Claude Code 源码](https://www.xuanyuancode.com/learn-claude-code)


---

> 作者:   
> URL: https://fengchen321.github.io/posts/ai/agent-%E6%9E%B6%E6%9E%84%E4%B8%8E%E5%B7%A5%E7%A8%8B/