{"data":{"slug":"xerrors-yuxi","name":"Yuxi","tagline":"A multi-tenant Agent Harness platform integrating knowledge base and knowledge graphs with LangChain + Vue + FastAPI.","github_url":"https://github.com/xerrors/Yuxi","owner":"xerrors","repo":"Yuxi","owner_avatar_url":"https://avatars.githubusercontent.com/u/35524243?v=4","primary_language":"Python","stars":6500,"forks":965,"topics":["agentic-rag","ai-agents","deepagents","docker","document-ai","fastapi","harness","kbqa","kgqa","knowledge-base","knowledge-graph","langgraph","llms","mcp","milvus","multi-agent","neo4j","rag","self-hosted","vue"],"archived":false,"github_pushed_at":"2026-08-18T11:44:36+00:00","maintenance_label":"Very active","stars_delta_30d":336,"url":"https://www.graphcanon.com/tools/xerrors-yuxi","markdown_url":"https://www.graphcanon.com/tools/xerrors-yuxi.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/xerrors-yuxi","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=xerrors-yuxi","description":"可私有部署的多租户知识智能体平台：统一 RAG、知识图谱、多智能体、MCP/Skills、沙盒与权限管理。Self-hosted knowledge agent platform for RAG, knowledge graphs and multi-agent workflows.","homepage_url":"https://xerrors.github.io/Yuxi/","license":"MIT","open_issues":81,"watchers":45,"ai_summary":"Yuxi is an agent-driven platform for managing knowledge bases and knowledge graphs using RAG retrieval, Milvus, and LangGraph. It supports multi-agent orchestration and integrates PDF mining, Neo4j, and more.","readme_excerpt":"<a href=\"https://trendshift.io/repositories/24335\" target=\"_blank\"><img src=\"https://trendshift.io/api/badge/repositories/24335\" alt=\"xerrors%2FYuxi | Trendshift\" style=\"width: 250px; height: 55px;\" width=\"250\" height=\"55\"/></a>\n\n[[项目主页]](https://xerrors.github.io/Yuxi/) · [[快速开始]](https://xerrors.github.io/Yuxi/intro/quick-start) · [[演示视频]](https://www.bilibili.com/video/BV1erE26iEgv/) · [[版本记录]](https://github.com/xerrors/Yuxi/releases) · [[English]](README.en.md)\n\n\n\n> 📢 作者为江南大学软件工程博士研究生，研究方向为 AI Agent、知识图谱与大模型应用，预计 2027年12月毕业，现寻求**实习/全职**机会。联系邮箱：wenjie.zhang@stu.jiangnan.edu.cn\n\n---\n\n## Yuxi 是什么\n\nYuxi（语析）是一个**可私有部署的多租户知识智能体平台**。把 **RAG 检索、Milvus 知识库内知识图谱、LangGraph 多智能体编排、MCP/Skills、沙盒工具与权限管理** 放进同一个工作台。\n\n管理员负责接入模型、建设知识库并配置用户与部门权限；用户在统一对话界面中调用知识、工具和子智能体，获得带来源引用、基于图谱上下文的推理以及可预览、可下载产物的回答。\n\n## 为什么选择 Yuxi\n\n- **知识与智能体真正协同**：知识库和知识图谱让 Agent 在运行时可检索、可引用的知识来源。\n- **从回答到任务交付**：Skills、MCP、工具、子智能体和沙盒文件系统共同支持长任务执行与产物交付。\n- **面向团队而非单用户 Demo**：提供多租户、用户/部门权限、统一模型配置和外部 API Key 集成。\n- **部署路径清晰**：Docker Compose 开箱即用。\n\n## 核心能力\n\n- **智能体运行时**：LangGraph、DeepAgents、SubAgents、Skills、MCP、Tools、中间件与异步 Worker。\n- **知识库与 RAG**：多格式解析、Embedding/Rerank、检索评估、来源引用和文件预览。\n- **知识图谱**：从知识库内容抽取实体关系，在 Milvus 与 Neo4j 中构建、检索并展示子图。\n- **沙盒与产物**：隔离文件系统，支持文本、图片、PDF、HTML 等产物落盘、预览和下载。\n- **平台治理**：用户与部门权限、模型供应商配置、API Key 调用、运行状态与评估能力。\n\n## 技术栈\n\n| 层 | 技术 |\n| --- | --- |\n| 前端 | Vue 3 · Vite · Pinia |\n| 后端 | FastAPI · LangGraph · ARQ (异步 worker) |\n| 存储 | PostgreSQL · Redis · MinIO · Milvus · Neo4j |\n| 文档解析 | MinerU · PaddleX · RapidOCR |\n| 部署 | Docker Compose |\n\n## 快速开始\n\n**前置要求**：已安装 [Docker](https://docs.docker.com/get-docker/) 与 Docker Compose，并准备至少一个兼容 OpenAI 接口的大模型 API。\n\n**1. 克隆代码并初始化**\n\n```bash\ngit clone --branch v0.7.1 --depth 1 https://github.com/xerrors/Yuxi.git\ncd Yuxi\n\n# Linux/macOS\n./scripts/init.sh\n\n# Windows PowerShell\n.\\scripts\\init.ps1\n```\n\n**2. 使用 Docker 启动**\n\n```bash\ndocker compose up --build\n```\n\n**3. 访问平台**\n\n等待启动完成后，浏览器打开 `http://localhost:5173`，使用初始化时生成的管理员账户登录即可。\n\n> 💡 不需要知识库 / 知识图谱等重依赖时，可使用 `make up-lite` 以 LITE 轻量模式启动，加快冷启动速度。更多部署说明见 [项目文档](https://xerrors.github.io/Yuxi)。\n\n详细配置、生产部署和故障排查请阅读[快速开始指南](https://xerrors.github.io/Yuxi/intro/quick-start)。最新开发动态见 [Changelog](https://xerrors.github.io/Yuxi/develop-guides/changelog)，规划中的能力见[开发路线图](https://xerrors.github.io/Yuxi/develop-guides/roadmap)。\n\n## 致谢\n\n本项目参考并引用了以下优秀开源项目，在此致以诚挚的感谢：\n\n- [LightRAG](https://github.com/HKUDS/LightRAG) - 早期版本曾参考其图谱构建与检索思路\n- [DeepAgents](https://github.com/langchain-ai/deepagents) - 直接引入作为深度智能体框架\n- [DeerFlow](https://github.com/bytedance/deer-flow) - 参考了其 Sandbox 智能体架构的实现思路\n- [RAGflow](https://github.com/infiniflow/ragflow) - 参考了其文档 Text Chunking 的分块策略\n- [LangGraph](https://github.com/langchain-ai/langgraph) - 多智能体编排框架，本项目的核心架构基础\n- [QwenPaw](https://github.com/agentscope-ai/QwenPaw) - 参考模型配置与个人文件区域设计\n\n## 参与贡献\n\n感谢所有贡献者的支持！\n\n<a href=\"https://github.com/xerrors/Yuxi/contributors\">\n  <img src=\"https://contrib.rocks/image?repo=xerrors/Yuxi&max=100&columns=10\" />\n</a>\n\n\n## Star History\n\n\n\n\n## 开源协议\n\n本项目采用 MIT 许可证 - 查看 [LICENSE](LICENSE) 文件了解详情。\n\nDocker Compose 引入的第三方组件（Neo4j 社区版 GPL-3.0、MinIO AGPL-3.0 等）保留各自原始许可证，部署与再分发边界见[生产部署指南](docs/advanced/deployment.md)。","github_created_at":"2024-07-05T10:57:35+00:00","created_at":"2026-07-07T17:37:48.480462+00:00","updated_at":"2026-08-18T18:01:16.165496+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"},{"slug":"data-retrieval","name":"Data & Retrieval","url":"https://www.graphcanon.com/categories/data-retrieval","markdown_url":"https://www.graphcanon.com/categories/data-retrieval.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/data-retrieval"}],"tags":[{"slug":"docker","name":"docker"},{"slug":"fastapi","name":"fastapi"},{"slug":"harness","name":"harness"},{"slug":"kbqa","name":"kbqa"},{"slug":"kgqa","name":"kgqa"},{"slug":"llms","name":"llms"},{"slug":"neo4j","name":"neo4j"},{"slug":"rag","name":"rag"}],"trust":{"provenance":{"is_fork":false,"github_id":824541776,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-08-18T18:01:14.605Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":2,"days_since_push":0,"last_release_at":"2026-07-17T15:50:37Z","stars_delta_30d":336,"open_issues_delta_30d":0},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:10:22.868Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-08-18T18:01:15.103Z"},"deploy":{"source":"dockerfile:docker-compose.yml","self_host":true,"observed_at":"2026-08-18T18:01:15.103Z","managed_saas":false},"languages":{"value":["python"],"source":"github.language","observed_at":"2026-08-18T18:01:15.103Z"},"has_docker":{"value":true,"source":"dockerfile:docker-compose.yml","observed_at":"2026-08-18T18:01:15.103Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-08-18T18:01:15.103Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["- Yuxi is ideal if you are looking for an agent-driven platform that supports RAG retrieval, Milvus, and integrates Neo4j for managing knowledge bases and graphs.","- If your project involves multi-agent orchestration with PDF mining capabilities, Yuxi offers a comprehensive solution within its framework."],"when_not_to_use":["- Avoid Yuxi if you need a platform that heavily relies on LightRAG as the original integration has been replaced by a custom Milvus-based implementation.","- Consider alternatives if you require advanced customization of document parsing beyond MinerU, PaddleX, and RapidOCR provided in Yuxi."],"source":"enrich:decision_facts","observed_at":"2026-07-15T10:38:48.291Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"Yuxi is a multi-tenant Agent Harness platform that integrates knowledge bases and knowledge graphs with LangChain + Vue + FastAPI."}]}}