# 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。

Repository: https://github.com/nageoffer/ragent
Canonical: https://ross.abutalabs.com/products/ragent
Homepage: https://nageoffer.com/ragent
Language: Java
License: Apache-2.0
License Family: permissive
Topics: agent, ai, rag, llm, springai, agentic-rag, mcp
Last push: 2026-08-25T16:15:37+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 85, longevity 20
- inputs: {"age_days": 286, "days_push": 8, "days_rel": 23, "gap_med": 55, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3779, forks 763 (observed 2026-08-28T04:08:18.388595+00:00)

## What it is
Ragent AI is a production-grade Java platform for building Agentic RAG applications, covering document ingestion, hybrid retrieval (vector, keyword, knowledge graph, web search), intent recognition, query rewriting, session memory, and MCP tool calling. It is also positioned as a hands-on learning project for Java backend developers transitioning into AI engineering.

## Use cases
- build an enterprise RAG chatbot over internal documents
- implement hybrid retrieval with vector, keyword, and knowledge graph recall
- add MCP tool calling to an LLM agent in Java
- manage conversation memory with token cost control
- learn how production RAG systems are engineered end to end
- add rate limiting and fair queuing in front of LLM services
- trace and evaluate RAG answers with source attribution

## When to choose
- you are a Java/Spring developer building a RAG or agent system in production
- you need a full pipeline from document ingestion to grounded answers with tracing
- you want a realistic, enterprise-level reference project for learning AI engineering

## When to avoid
- you need a Python-based RAG stack like LangChain or LlamaIndex
- you only need a lightweight demo or a simple vector search wrapper
- you want a managed SaaS RAG service rather than a self-hosted platform

## Facets
- artifact type: application
- maturity: active
- function: rag, agent-framework, mcp, search-engine, chatbot, llm-inference, prompt-engineering, web-framework
- domain: large-language-models, chatbots, developer-tools, education
- platform: jvm, self-hosted
- tags: agentic-rag, spring-ai, hybrid-retrieval, knowledge-base, rerank, session-memory, interview-preparation, enterprise, retrieval-augmented-generation, ai-agents, natural-language-processing, web-server, docker

## Member repositories
- nageoffer/ragent (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:18.388595+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:27:46.878975+00:00, confidence not recorded.
  - readme: https://github.com/nageoffer/ragent (fetched 2026-08-28T04:08:18.388595+00:00, sha 5d53f1e3ec39)
  - homepage: https://nageoffer.com/ragent (fetched 2026-08-29T09:22:35.789576+00:00, sha edc52d08efcb)
  - site_page: https://nageoffer.com/docs (fetched 2026-08-29T09:22:35.792794+00:00, sha 2085d089b778)
  - site_page: https://nageoffer.com/ragent/faq (fetched 2026-08-29T09:22:35.794449+00:00, sha f7f3746c2f3e)
- Data as of 2026-08-30T08:39:29.467469+00:00.