Whliu Manus
What is Whliu Manus
whliu-Manus is an advanced AI project that implements large model dialogues using Spring AI, featuring RAG (Retrieval-Augmented Generation), custom interceptors, dialogue persistence, structured output converters, tool invocation, and MCP.
Use cases
Use cases include automated customer support agents, intelligent personal assistants, document retrieval systems, and applications needing advanced dialogue management and tool integration.
How to use
To use whliu-Manus, set up the environment with Spring AI, SpringBoot 3, and JDK21. You can interact with the intelligent agent through the MCP server, which allows the agent to autonomously call local tools and generate results based on its reasoning.
Key features
Key features include custom interceptors, ChatMemory definitions, structured output converters, dialogue memory persistence, local and online RAG, vector databases, document retrievers, context query enhancers, function call development, and custom Manus agents.
Where to use
whliu-Manus can be used in various fields including AI research, software development, customer support automation, and any application requiring intelligent dialogue systems and data retrieval.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is Whliu Manus
whliu-Manus is an advanced AI project that implements large model dialogues using Spring AI, featuring RAG (Retrieval-Augmented Generation), custom interceptors, dialogue persistence, structured output converters, tool invocation, and MCP.
Use cases
Use cases include automated customer support agents, intelligent personal assistants, document retrieval systems, and applications needing advanced dialogue management and tool integration.
How to use
To use whliu-Manus, set up the environment with Spring AI, SpringBoot 3, and JDK21. You can interact with the intelligent agent through the MCP server, which allows the agent to autonomously call local tools and generate results based on its reasoning.
Key features
Key features include custom interceptors, ChatMemory definitions, structured output converters, dialogue memory persistence, local and online RAG, vector databases, document retrievers, context query enhancers, function call development, and custom Manus agents.
Where to use
whliu-Manus can be used in various fields including AI research, software development, customer support automation, and any application requiring intelligent dialogue systems and data retrieval.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
介绍
炫技项目，手搓Manus智能体、MCP server、MCP Client、离线RAG、Function Call等，用代码揭开LLM开发的真面目。
环境
- Spring AI
- SpringBoot 3
- JDK21
智能体实现效果
可以自主调用本地实现的MCP服务中的工具，自主思考，最终给出结果
执行过程，智能体可以自主思考调用什么工具，并根据工具调用返回的结果来决定下一步的策略：
生成PDF结果（markdown格式，图片加载有问题）：
图片链接可以打开：
目前已完成功能
- 自定义拦截器（Advisor）
- 违禁词校验Advisor
- 专用模板类，从文件加载模板
- 自定义ChatMemory
- 结构化输出转换器
- 对话记忆持久化
- 本地RAG
- 在线RAG
- 向量数据库
- 文档检索器
- 上下文查询增强器
- function call开发
- MCP开发（sse和stdio）
- 自定义Manus智能体
TODO
- 权限校验
- 自定义对话记忆，持久化对话到MySQL、Redis中
- 自定义DocumentReader，读取Github仓库信息
- 自定义QueryTransformer查询转换器，使用第三方翻译API代替大模型翻译
- 实现基于向量数据库和其他数据库的混合检索（MySQL、Redis、ES）
- 自动发送邮件tool
- 优化PDF生成，本地存储图片
- 手动控制ToolCallingManager执行流程
- 实现env环境变量传递参数，多agent工作流
Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.
