[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-9011":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":15,"stars7d":15,"stars30d":14,"stars90d":15,"forks30d":15,"starsTrendScore":15,"compositeScore":16,"rankGlobal":9,"rankLanguage":9,"license":17,"archived":18,"fork":18,"defaultBranch":19,"hasWiki":18,"hasPages":18,"topics":20,"createdAt":9,"pushedAt":9,"updatedAt":21,"readmeContent":22,"aiSummary":23,"trendingCount":15,"starSnapshotCount":15,"syncStatus":24,"lastSyncTime":25,"discoverSource":26},9011,"x-agent","wenge-research\u002Fx-agent","wenge-research","智川x-agent",null,"Vue",1087,102,68,1,0,19.04,"GNU General Public License v3.0",false,"main",[],"2026-06-12 02:02:01","\u003Cp align=\"center\">\u003Cimg src= \"https:\u002F\u002Fdibrain.wenge.com\u002Fwg-agent-manage-uat\u002Fstatic\u002Fimg\u002Flogo-new.81fbf2b9.png\" alt=\"x-agent\" width=\"200\" \u002F>\u003C\u002Fp>\n\u003Ch3 align=\"center\">Open-source platform for building enterprise-grade agents\u003C\u002Fh3>\n\u003Ch3 align=\"center\">​All-in-One Enterprise Agent Development Platform\u003C\u002Fh3>\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fwww.gnu.org\u002Flicenses\u002Fgpl-3.0.html#license-text\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-GPL3.0-blue\" alt=\"License: GPL v3\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002F1Panel-dev\u002Fmaxkb\u002Freleases\u002Flatest\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Frelease-V1.0.1-blue\" alt=\"Latest release\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Chr\u002F>\n\u003Cp>\n智川X-Agent是中科闻歌推出的一站式企业智能体开发平台，帮助企业零代码快速构建AI应用。智川X-Agent基于封装大模型、知识库、工作流等复杂技术模块为可视化组件，用户通过简单的拖拽和配置可搭建符合业务需求的AI应用。智川X-Agent平台支持多种大模型（如雅意、文心一言等），提供知识库管理、工作流编排、应用发布等功能，满足政务、金融、媒体等多行业需求，助力企业实现AI应用的极速落地与高效迭代，加速AI普惠化。\n\u003C\u002Fp>\n\u003Chr\u002F>\n## 快速开始\n\n通过docker 镜像启动 x-agent:\n\n```bash\ndocker volume create agent-x-data\ndocker volume create agent-x-cicd\ndocker pull ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:agent-x_no_bge_250815_05\ndocker run  -d --restart=always -p 80:80 -p 443:443 -p 8848:8848 -p 3306:3306 -p 6379:6379 -p 9200:9200 -p 9000:9000 -p 9001:9001 -e IP_ADDR=\"127.0.0.1:80\" -v agent-x-data:\u002Fu01\u002Fisi -v agent-x-cicd:\u002Fapp\u002Fagent\u002Fserver  --name agent-x  ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:agent-x_no_bge_250815_05\n```\n访问管理后台 `http:\u002F\u002F127.0.0.1:80\u002Fwg-agent-manage\u002F#\u002Fappmanage` \n\n- 账号: agent-x\n- 密码: 04p9xa0gAE*%&Op8\n\n\n## 访问 nacos\n> http:\u002F\u002F127.0.0.1:8848\u002Fnacos\u002F\n> - 账号: nacos\n> - 密码: k2j210w5CKKO!&Wh0\n\n## 连接数据 mysql\n> 127.0.0.1:3306\n> - 账号: root\n> - 密码: 2ievD%GBA6\n> - 主库：smart_customer_agent\n\n## 访问minio\n> http:\u002F\u002F127.0.0.1:9000\n> - 账号: admin\n> - 密码: 6838BHE%%C472\n\n\n## 启动算法服务\n\n```bash\ndocker pull ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\nmkdir -p \u002Fu01\u002Fisi\u002Fcode_sdk\n#下载 算法配置文件包并上传到服务器指定目录，配置文件包在目录：\u002Fconfig\u002FAgent_X.zip\nunzip Agent_X.zip\n```\n\n### 1.启动向量模型\n```bash\ndocker run -d -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FEmbedding_model\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FEmbedding_model\u002Fmain.py:\u002Fapp\u002Fmain.py  -p 10822:8080 ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n```\n\n### 2.工作流代码节点\n```bash\ndocker run -d -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FCode_node\u002Fmain.py:\u002Fapp\u002Fmain.py  -p 1216:8080 ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n```\n\n### 3.智能问数（NL2SQL）\n```bash\ndocker run -d -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FNl2sql\u002Fconfig.yaml:\u002Fapp\u002Fconfig.yaml -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FNl2sql\u002Fmain.py:\u002Fapp\u002Fmain.py  -p 1025:8080 ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n```\n\n### 4.MCP\n```bash\ndocker run -d -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FMCP\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FMCP\u002Fmain.py:\u002Fapp\u002Fmain.py  -p 4011:8080  ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n```\n\n### 5.本地自定义安装 mcp\n```bash\ndocker run -d --name algorithm-local-mcp --net host -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FAuto_mcp\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FAuto_mcp\u002Fmain.py:\u002Fapp\u002Fmain.py  -v  \u002Fu01\u002Fisi\u002Fcode_sdk\u002FAuto_mcp\u002Fmcp_file:\u002Fapp\u002Fmcp_file ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n```\n\n### 6.MCP_nl2sql\n```bash\ndocker run -d -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002Fmcp_sql\u002Fconfig.yaml:\u002Fapp\u002Fconfig.yaml -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002Fmcp_sql\u002Fmain.py:\u002Fapp\u002Fmain.py  -p 4016:8080 ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n```\n\n```bash\ndocker pull ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_chrome_v2\n```\n\n### 7.单网页内容爬取\n```bash\ndocker run -d -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FURL_analysis\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FURL_analysis\u002Fmain.py:\u002Fapp\u002Fmain.py  -p 9007:8080 ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_chrome_v2\n```\n\n### 8.网页截图\n```bash\ndocker run -d -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FURL_to_img\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml -v \u002Fu01\u002Fisi\u002Fcode_sdk\u002FURL_to_img\u002Fmain.py:\u002Fapp\u002Fmain.py  -p 5028:8080 ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_chrome_v2\n```\n\n## yayi算法\n### 9.重排序服务\n```bash\ndocker pull ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:reranker.v1\ndocker run -d -p 9098:8080 --restart=always --name reranker ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:reranker.v1\n```\n\n### 10.文档智能解析\n```bash\ndocker pull ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:contentparse.v4.7\ndocker run -d -p 9099:8080 --restart=always  --name content_parse ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:contentparse.v4.7\n```\n\n### 11.文档切片\n```bash\ndocker pull ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:doc_answer_noes_nosql.v1.2.8-build2503143-encrypted\ndocker run -d -p 9097:8080 --restart=always  --name yayi-plugin-doc-answer-250623 ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:doc_answer_noes_nosql.v1.2.8-build2503143-encrypted\n```\n\n---\n> **注意**：如果默认服务的 ip 和端口有变动，请调整对应的配置项！\n### nacos配置项\n```yaml\nappframe:\n  yayi:\n    # 10.文档智能解析\n    contentparsingnewversion:\n      uri: http:\u002F\u002F172.17.0.0.1:9099\u002Fanalysis\n    # 9.重排序服务\n    rearrange:\n      uri: http:\u002F\u002F172.17.0.0.1:9098\u002Fanalysis\n    # 11.文档切片\n    knowledgesplit:\n      uri: http:\u002F\u002F172.17.0.1:9097\u002Fanalysis\n# 网页快照\nscreenshot:\n  # 图片上传\n  uploadUrl: http:\u002F\u002F172.17.0.1:80\u002Fsmart-agent-api\u002Fwos\u002Ffile\u002Fupload\n  # 8.网页截图\n  api: http:\u002F\u002F172.17.0.1:5028\u002Fcapture-screenshot\n\n# mcp服务api\nmcp:\n  # 4.MCP\n  serviceApi: http:\u002F\u002F172.17.0.1:4011\u002Fservice\n  # 4.MCP\n  queryApi: http:\u002F\u002F172.17.0.1:4011\u002Fquery\n  # 5.本地自定义安装 mcp\n  buildMcpApi: http:\u002F\u002F172.17.0.1:4011\u002Fdeploy_service\n  # 3.智能问数（NL2SQL）\n  textToSqlSse: http:\u002F\u002F172.17.0.1:1025\u002Fget_answer_text2sql\n\nworkflow:\n  default:\n    # 2.工作流代码节点\n    codeApi: http:\u002F\u002F172.17.0.1:1216\u002Fexecute\n    # 开始节点的改写时使用的大模型 id（llm_info表的 model_id字段）\n    startRewriteModelId: 87026c3464664ad49a8b622ec719fa70\n```\n### MYSQL配置\n```mysql\n-- 1.启动向量模型\nuse smart_customer_agent;\nupdate smart_customer_agent.dense_vector set uri='http:\u002F\u002F172.17.0.1:10822\u002Fanalysis' where code = 'local_bge_768';\n```\n\n## 技术栈\n\n- 前端： vue\n-  后端： java\n-  算法： python\n-  数据库： mysql\n- 中间件：elasticsearch,redis,minio,nginx\n\n### docker-compose.yml\n```yaml\n\nversion: '3.8'\nservices:\n  # ========== 主服务：x-agent ==========\n  agent-x:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:agent-x_no_bge_250815_05\n    container_name: agent-x\n    restart: always\n    ports:\n      - \"80:80\"     # 替换为浏览器将要访问的端口，与IP_ADDR参数的端口一致\n      - \"443:443\"\t# https\n      - \"8848:8848\" # Nacos\n      - \"3306:3306\" # MySQL\n      - \"6379:6379\" # Redis\n      - \"9200:9200\" # Elasticsearch\n      - \"9000:9000\" # MinIO\n      - \"9001:9001\" # MinIO\n    environment:\n      - IP_ADDR=127.0.0.1:80 # 替换为浏览器将要访问的 ip 和端口\n    volumes:\n      - agent-x-data:\u002Fu01\u002Fisi\n      - agent-x-cicd:\u002Fapp\u002Fagent\u002Fserver\n\n  # ========== 算法服务 ==========\n  # 1. 向量模型\n  algorithm-vector:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n    container_name: algorithm-vector\n    restart: always\n    ports:\n      - \"10822:8080\"  # 默认端口10822，如果默认端口10822有变动，请修改 mysql 的配置项:use smart_customer_agent; update smart_customer_agent.dense_vector set uri='http:\u002F\u002F172.17.0.1:10822\u002Fanalysis' where code = 'local_bge_768';\n    volumes:\n      - .\u002Fcode_sdk\u002FEmbedding_model\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml\n      - .\u002Fcode_sdk\u002FEmbedding_model\u002Fmain.py:\u002Fapp\u002Fmain.py\n\n  # 2. 工作流代码节点\n  algorithm-code-node:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n    container_name: algorithm-code-node\n    restart: always\n    ports:\n      - \"1216:8080\"  #默认端口1216，如果默认端口1216有变动，请修改 nacos 的配置项: workflow.default.codeApi: http:\u002F\u002F172.17.0.1:1216\u002Fexecute\n    volumes:\n      - .\u002Fcode_sdk\u002FCode_node\u002Fmain.py:\u002Fapp\u002Fmain.py\n\n  # 3. 智能问数（NL2SQL）\n  algorithm-nl2sql:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n    container_name: algorithm-nl2sql\n    restart: always\n    ports:\n      - \"1025:8080\" #默认端口1025，如果默认端口1025有变动，请修改 nacos 的配置项: mcp.textToSqlSse: http:\u002F\u002F172.17.0.1:1025\u002Fget_answer_text2sql\n    volumes:\n      - .\u002Fcode_sdk\u002FNl2sql\u002Fconfig.yaml:\u002Fapp\u002Fconfig.yaml\n      - .\u002Fcode_sdk\u002FNl2sql\u002Fmain.py:\u002Fapp\u002Fmain.py\n\n  # 4. MCP\n  algorithm-mcp:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n    container_name: algorithm-mcp\n    restart: always\n    ports:\n      - \"4011:8080\" #默认端口4011，如果默认端口4011有变动，请修改 nacos 的配置项: mcp.buildMcpApi: http:\u002F\u002F172.17.0.1:4011\u002Fdeploy_service\n    volumes:\n      - .\u002Fcode_sdk\u002FMCP\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml\n      - .\u002Fcode_sdk\u002FMCP\u002Fmain.py:\u002Fapp\u002Fmain.py\n\n  # 5. 本地自定义 MCP (使用 host 网络)\n  algorithm-local-mcp:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_v2\n    container_name: algorithm-local-mcp\n    restart: always\n    network_mode: host  # 使用宿主机网络\n    volumes:\n      - .\u002Fcode_sdk\u002FAuto_mcp\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml\n      - .\u002Fcode_sdk\u002FAuto_mcp\u002Fmain.py:\u002Fapp\u002Fmain.py\n      - .\u002Fcode_sdk\u002FAuto_mcp\u002Fmcp_file:\u002Fapp\u002Fmcp_file\n\n  # 8. 网页截图\n  algorithm-url-to-img:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:algorithm_chrome_v2\n    container_name: algorithm-url-to-img\n    restart: always\n    ports:\n      - \"5028:8080\" #默认端口5028，如果默认端口5028有变动，请修改 nacos 的配置项: screenshot.api: http:\u002F\u002F172.17.0.1:5028\u002Fcapture-screenshot\n    volumes:\n      - .\u002Fcode_sdk\u002FURL_to_img\u002Fconfig.yml:\u002Fapp\u002Fconfig.yml\n      - .\u002Fcode_sdk\u002FURL_to_img\u002Fmain.py:\u002Fapp\u002Fmain.py\n\n  # 9. 重排序服务 (yayi)\n  algorithm-reranker:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:reranker.v1\n    container_name: algorithm-reranker\n    restart: always\n    ports:\n      - \"9098:8080\" #默认端口9098，如果默认端口9098有变动，请修改 nacos 的配置项: appframe.yayi.rearrange.uri: http:\u002F\u002F172.17.0.0.1:9098\u002Fanalysis\n\n  # 10. 文档智能解析 (yayi)\n  algorithm-content-parse:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:contentparse.v4.7\n    container_name: algorithm-content-parse\n    restart: always\n    ports:\n      - \"9099:8080\" #默认端口9099，如果默认端口9099有变动，请修改 nacos 的配置项: appframe.yayi.contentparsingnewversion.uri: http:\u002F\u002F172.17.0.0.1:9099\u002Fanalysis\n\n  doc_answer_noes:\n    image: ccr.ccs.tencentyun.com\u002Fwenge\u002Fagent-x:doc_answer_noes_nosql.v1.2.8-build2503143-encrypted\n    container_name: doc_answer_noes\n    restart: always\n    ports:\n      - \"9097:8080\" #默认端口9097，如果默认端口9097有变动，请修改 nacos 的配置项: appframe.yayi.knowledgesplit.uri: http:\u002F\u002F172.17.0.0.1:9097\u002Fanalysis\n\n# ========== 数据卷定义 ==========\nvolumes:\n  agent-x-data:\n    external: true  # 对应你之前创建的 docker volume create agent-x-data\n  agent-x-cicd:\n    external: true  # 对应你之前创建的 docker volume create agent-x-cicd\n\n```\n\n\n### 相关脚本\n```shell\n# 启动所有服务\ndocker-compose up -d\n# 查看运行状态\ndocker-compose ps\n# 查看日志\ndocker-compose logs -f [服务名]\n```\n\n## 联系工作人员\n\u003Cimg src=\".\u002Fworker.png\" alt=\"图片描述\" width=\"200\" \u002F>\n","智川X-Agent是由中科闻歌推出的一站式企业智能体开发平台，旨在帮助企业无需编写代码即可快速构建AI应用。该平台基于Vue框架开发，通过将复杂的大模型、知识库和工作流等技术封装成可视化组件，用户仅需简单拖拽与配置即可完成符合业务需求的应用搭建。智川X-Agent支持多种大模型（如雅意、文心一言），并提供包括知识库管理、工作流编排以及应用发布在内的多项功能，特别适用于政务、金融及媒体等行业场景，助力这些领域的AI应用实现快速部署与迭代升级。",2,"2026-06-11 03:20:47","top_language"]