Yu AI Agent (YuManus)

Overview#

Yu AI Agent (YuManus) is a hands-on AI development tutorial project for programmers, built on Spring Boot 3.4.4 and Java 21, integrating the Spring AI framework.

Core Features#

Model Integration#

• Alibaba Cloud Bailian/DashScope: Via Spring AI Alibaba (DashScope SDK 2.19.1)
• Ollama Local Models: Support for edge-side model deployment
• LangChain4J Integration: Alternative implementation path
• Unified Multi-Model Access: Flexible switching between cloud and local models

AI Capabilities#

• Prompt Engineering: Prompt design and optimization, context management
• RAG (Retrieval-Augmented Generation): Vector database integration, similarity search
• Tool Calling: External tool and service invocation
• MCP Protocol: Model Context Protocol implementation for AI interaction context
• ReAct Agent: Autonomous planning agent (Reasoning + Acting pattern)

Use Cases#

1. AI Love Master conversational bot
2. Autonomous planning agent YuManus
3. Enterprise AI application prototyping
4. Java developer AI skills learning

Requirements#

• JDK 21 (Required)
• Maven 3.x or Maven Wrapper

Quick Start#

# Clone repository
git clone https://github.com/hfgwygey/yu-ai-agent.git

# Enter directory
cd yu-ai-agent

# Install dependencies
mvn install

# Run application
mvn spring-boot:run

Access Points#

• Application: http://localhost:8080
• API Docs: http://localhost:8080/doc.html (based on Knife4j convention)

Key Dependencies#

• Spring Boot 3.4.4
• Spring AI 1.0.0-M6
• spring-ai-alibaba-starter 1.0.0-M6.1
• spring-ai-ollama-spring-boot-starter 1.0.0-M6
• langchain4j-community-dashscope 1.0.0-beta2
• Knife4j OpenAPI3 4.4.0

Configuration#

Requires Spring Milestones repository for Spring AI dependencies, and configuration of Alibaba Cloud DashScope API key or Ollama service URL in application.yml.

Project Status#

Version 0.0.1-SNAPSHOT, early development stage, suitable for learning and technical exploration.