GH Repository · nageoffer
ragent
企业级 Agentic RAG 智能体 — 全链路覆盖文档解析、多路检索、意图识别、问题重写、会话记忆、MCP 工具调用与深度思考。面向真实业务场景，从 0 到 1 完整工程实现。
- stars
- 4,035
- 30-day movement
- +5117/day
- Related entries
- 60
- Connections
- 2
java/mavenJavamcpagentic-ragagent-appspringairagllmaiagent
ragent is an enterprise Agentic RAG agent built in Java with Spring AI, covering document parsing, multi-route retrieval, intent recognition, query rewriting, session memory, MCP tool calls, and deep thinking. It is a complete from-zero engineering implementation aimed at real business scenarios.
Reach for it when you need a production-oriented Java reference for building an agentic RAG pipeline end to end.
Use it to
- Build an agentic RAG service in Java with Spring AI
- Study a full pipeline from document parsing to multi-route retrieval
- Add intent recognition and query rewriting to a RAG flow
- Integrate MCP tool calls and session memory into an agent
- Use the shipped claude directory as a starting configuration
For Java developers building enterprise RAG agents
- Role
- agent-app
- Language
- Java
- Licence
- Apache-2.0
- Forks
- 817
- Open issues
- 33
- Last push
- 2026-09-17
- Latest release
- 1.0.0 · 2026-06-16
topicsagentragllmspringaimcpjava