解释 BK-CI 流水线或创作流的 Model JSON,帮助 AI 理解 Model -> Stage -> Container(Job) -> Element 的含义、启动参数、矩阵、Finally、审核与常见插件语义。用户贴出流水线 JSON、创作流 JSON、编排数据,或询问某个 Stage/Job/插件/atomCode/stepId 含义时使用。
72
89%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Passed
No findings from the security scan
Model JSONatomCode、elementId、stepId、jobId 等信息,要求说明其在编排中的含义Model -> Stage -> Container -> Element 四层结构。Stage 里的 TriggerContainer.params 优先解释为启动参数入口;触发器插件本身通常不用当业务步骤展开。trigger、vmBuild、normal,再判断插件类型:linuxScript、windowsScript、marketBuild、marketBuildLess、subPipelineCall、manualReviewUserTask 等。@type、atomCode、name、data.input、data.output 和 additionalOptions。atomCode 时,不臆测功能;明确标注“需结合插件文档确认”,并列出关键输入输出字段。runCondition 判断;Element / Job / Stage 三层各有独立枚举,不要混用。reference/。containerHashId、jobId、element.id、stepId 语义不同,不能混用status、startEpoch、elapsed、executeCount、timeCost 等运行态字段默认忽略,除非用户明确问运行状态additionalOptions.enable=false、jobControlOption.enable=false、stageControlOption.enable=false 要明确标注为已禁用PASSWORD 类型只标注为敏感,不回显默认值atomCode 不看 data.input / data.output,导致插件理解失真reference/1-json-analysis-playbook.mdreference/2-common-plugin-and-container-hints.mdreference/3-classtype-and-enum-catalog.mdmanaging-devops-pipeline 或构建 Agent 的编排查询工具(获取流水线编排摘要/获取流水线编排)pipeline-model-architecture94743cb
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.