[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83830-en":3,"doc-seo-83830-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},83830,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","LLM驱动的网络安全CI/CD工作流智能分析管线","CI/CD工作流被视为可执行的运行策略：它们决定构建、测试、发布与部署做什么，并塑造维护者与交付基础设施的交互方式。该工作面向网络系统工程的可观测性问题，指出仅靠工作流阶段标签无法判断脆弱性、生态异常性及优先级修订价值。作者提出基于LLM的CI/CD分析管线：仓库增强、反模式检测、阶段挖掘与推荐生成。结果覆盖59,550个星标仓库、75,201个工作流与434,769条反模式发现，阶段使用在语言间显著差异，few-shot在生成可校验建议上表现最佳。","LLM-Driven CI/CD Workflow Intelligence for Cyber Systems  \nEngineering  \nBonan Shen 1 , Jiazhou Gao2 Tao Ning3 , Wei-Jung Huang4 and Xin Liu5  \narXiv :2607 .04579v 1 [ cs . SE] 6 Jul 2026  \nAbstract—CI/CD workflows have become executable operational policy: they decide what gets built, tested, released, and deployed, and they mediate how maintainers interact with delivery infrastructure. That makes them an important measurement point for cyber-systems engineering. Recent large language model (LLM) work shows that workflow stages can be recognized directly from configuration files, but stage labels alone do not tell us whether a workflow is brittle, unusual for its ecosystem, or worth revising first. We present an LLM-based CI/CD analysis pipeline that combines repository enrichment, anti-pattern detection, stage mining, and recommendation generation over a large GitHub corpus. Starting from 59,550 repositories with at least 1,000 stars, we identify 34,225 projects with CI/CD and collect 127,559 configuration files. Across 75,201 analyzed workflows, the anti-pattern detector reports 434,769 findings, dominated by reliability and maintainability issues. Across 59,906 configurations, stage usage differs significantly by language (χ2 = 4168 .88, p \u003C 0.001, Cramer’s V = 0 .063), and domain analysis shows distinct operational profiles, including higher release and cache usage in mobile projects. For repository-level recommendation generation, fewshot prompting performs best overall, averaging 8.25 recommendations per repository with 96.1% YAML-valid snippets. Taken together, the results argue for CI/CD observability that combines diagnosis, context, and human review rather than treating workflow mining as a stage-classification problem alone.  \nI. INTRODUCTION  \nCI/CD configuration files do more than automate builds. They decide which checks gate a merge, how failures are bounded, when artifacts are published, and what paths lead to deployment. In other words, they encode operational policy. For software-intensive systems, that policy has direct consequences for reliability, security, release cadence, and infrastructure cost. Recent work has shown that LLMs can recover common CI/CD stages directly from raw workflow files at scale [1] . Useful as that result is, it still leaves the main engineering questions open: Which workflows look fragile? What counts as normal for a given ecosystem? Which fixes are worth a maintainer’s time?  \nWe study those questions by treating CI/CD workflows as analyzable infrastructure rather than passive configuration text. Our pipeline begins with a large GitHub corpus, enriches repositories with language and domain metadata, scans workflows for anti-patterns and stage structure, and then generates repository-level recommendations with sev-  \n1 Independent Researcher [shenbonan2@gmail.com](shenbonan2@gmail.com)  \n[2](2 Independent Researcher gjz140103@gmail.com)[ Independent Researcher](2 Independent Researcher gjz140103@gmail.com)[ gjz140103@gmail.com](2 Independent Researcher gjz140103@gmail.com)  \n[3](3 Independent Researcher ntgd1102@gmail.com)[ Independent Researcher](3 Independent Researcher ntgd1102@gmail.com)[ ntgd1102@gmail.com](3 Independent Researcher ntgd1102@gmail.com)  \n[4](4 Independent Researcher william.wj.huang@gmail.com)[ Independent Researcher](4 Independent Researcher william.wj.huang@gmail.com)[ william.wj.huang@gmail.com](4 Independent Researcher william.wj.huang@gmail.com)  \n5 Independent Researcher [iamxinliu@gmail.com](iamxinliu@gmail.com)  \n| Prompting comparison validity analysis | Language significance domain stage profiles |\n| --- | --- |\n\nFig. 1. Study overview. The artifact extends CI/CD analysis from coarse stage recognition to a connected pipeline that enriches repositories, mines anti-patterns and stages, and generates actionable repair suggestions.  \neral prompting strategies. The goal is not to replace maintainers with an autonomous fixer. The goal is to give them a cl","cbCaisNejrWh621X","https://ap.wps.com/l/cbCaisNejrWh621X","pdf",215893,1,6,"English","en",105,"# Abstract\n# I. INTRODUCTION","[{\"question\":\"为什么CI/CD工作流对网络系统工程很关键？\",\"answer\":\"CI/CD工作流承载可执行的运行策略，决定构建、测试、发布与部署规则，并影响可靠性、安全性、发布节奏与基础设施成本等关键工程结果。\"},{\"question\":\"文中提出的LLM分析管线包含哪些主要步骤？\",\"answer\":\"管线包含仓库增强（元数据补全）、反模式检测、阶段挖掘，以及对单个仓库生成可操作的推荐建议。\"},{\"question\":\"在自动生成CI/CD建议方面，哪种提示策略效果最好？\",\"answer\":\"few-shot prompting整体表现最佳，平均每个仓库生成8.25条建议，且96.1%的片段满足YAML语法可用性；迭代式提示会更偏向输出关键问题。\"}]",1784190781,15,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"llm-driven-cicd-workflow-intelligence-for-cyber-systems-engineering","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/llm-driven-cicd-workflow-intelligence-for-cyber-systems-engineering/83830/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"为什么CI/CD工作流对网络系统工程很关键？","Question",{"text":75,"@type":76},"CI/CD工作流承载可执行的运行策略，决定构建、测试、发布与部署规则，并影响可靠性、安全性、发布节奏与基础设施成本等关键工程结果。","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"文中提出的LLM分析管线包含哪些主要步骤？",{"text":80,"@type":76},"管线包含仓库增强（元数据补全）、反模式检测、阶段挖掘，以及对单个仓库生成可操作的推荐建议。",{"name":82,"@type":73,"acceptedAnswer":83},"在自动生成CI/CD建议方面，哪种提示策略效果最好？",{"text":84,"@type":76},"few-shot prompting整体表现最佳，平均每个仓库生成8.25条建议，且96.1%的片段满足YAML语法可用性；迭代式提示会更偏向输出关键问题。","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]