pipeline-model-interpreter

Interprets BK-CI pipeline Model JSON into Stage, Container, and Element semantics.

2.5k|524|Updated May 29, 2019
One-click install
npx skills add https://github.com/TencentBlueKing/bk-ci --skill pipeline-model-interpreter
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pipeline-model-interpreter
Source: https://github.com/TencentBlueKing/bk-ci/tree/main/ai/skills/pipeline-model-interpreter
Command: npx skills add https://github.com/TencentBlueKing/bk-ci --skill pipeline-model-interpreter

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

BK-CI pipeline and authoring-flow configurations are stored as deeply nested Model JSON (Model -> Stage -> Container -> Element), which is hard to read directly. This Skill helps an AI explain what a pasted pipeline JSON actually defines: startup parameters, job types, plugin behavior, matrix expansion, Finally stages, and review gates.

Core Features & Use Cases

  • Model Structure Decoding: Restores raw JSON into the four-layer Model -> Stage -> Container -> Element hierarchy and explains each layer's role.
  • Plugin Semantics: Interprets built-in elements (linuxScript, windowsScript, CODE_GIT, manualReviewUserTask, subPipelineCall) and store plugins (marketBuild/marketBuildLess) via atomCode, data.input, and data.output, with conservative handling of unknown atomCodes.
  • Control Flow Explanation: Clarifies runCondition enums at Stage/Job/Element levels, matrix expansion, mutex groups, checkIn/checkOut reviews, and disabled nodes.
  • Use Case: A user pastes a pipeline JSON and asks "what does this pipeline do?" The Skill explains the startup parameters from TriggerContainer.params, walks through each Stage and Job, and summarizes the overall flow and likely outputs.

Quick Start

Paste your BK-CI pipeline Model JSON and ask the AI to explain what this pipeline does, including its startup parameters, stages, jobs, and plugins.

Frequently Asked Questions about pipeline-model-interpreter

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I understand a BK-CI pipeline Model JSON?

Restore the JSON into the four-layer Model -> Stage -> Container -> Element structure first. Then read TriggerContainer.params for startup parameters, classify each container as trigger, vmBuild, or normal, and interpret each element by its @type, atomCode, data.input, and data.output.

What is the difference between containerHashId, jobId, and stepId in BK-CI?

containerHashId is the stable system identifier of a Job container, jobId is a user-defined Job identifier, and stepId is a user-defined step identifier on an element. They locate nodes at different layers and must not be treated as the same ID.

How are BK-CI store plugins (marketBuild) interpreted from JSON?

Read atomCode to identify the plugin, then examine data.input for what it does and data.output plus namespace for what it produces. For unknown atomCodes, state that the function needs confirmation against plugin documentation rather than guessing.

Does the runCondition enum work the same at Stage, Job, and Element levels?

No. Stage, Job, and Element each have independent runCondition enums with different literals, such as AFTER_LAST_FINISHED for stages, PREVIOUS_STAGE_SUCCESS for jobs, and PRE_TASK_SUCCESS for elements. They must not be mixed when explaining execution timing.

When should this Skill not be used for pipeline questions?

It does not apply to modifying model code, debugging runtime build failures or execution logs, designing summary output templates, or understanding YAML conversion and version compatibility internals. Those scenarios need other tools or skills.