一键导入
dew-serving-design
Design how Gold models, metrics, and data products are served to consumers with trust, caveats, contracts, and feedback loop.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Design how Gold models, metrics, and data products are served to consumers with trust, caveats, contracts, and feedback loop.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Clarify business decision, data consumers, stakeholder context, and decision workflow before KPI and source design.
Review implemented data engineering story for AC compliance, DQ evidence, grain, lineage, operational behavior, and caveats.
Create a ready-for-dev data engineering story with context, evidence requirements, acceptance criteria, tests, and Definition of Done.
Convert approved DEW designs into data engineering epics, story map, dependencies, and implementation backlog.
Authors and updates customization overrides for installed DEW skills.
Create evidence-grounded data architecture from requirement gate, KPI feasibility, source validation, and approved caveats.
| name | dew-serving-design |
| description | Design how Gold models, metrics, and data products are served to consumers with trust, caveats, contracts, and feedback loop. |
Goal: Define how the data product is served to consumers.
Your Role: You are a data product serving facilitator.
You translate Gold models, KPI definitions, DQ rules, and consumer needs into a clear serving specification.
You may recommend, but you must not decide.
This workflow uses step-file architecture.
Rules:
HALT-15 — Serving Consumer Unclear.HALT-14 — DQ Rules Missing.HALT-16 — Security Privacy Unreviewed.{skill-root} resolves to this skill's installed directory.{project-root}-prefixed paths resolve from the project working directory.{workflow.<name>} resolves to fields in customize.toml's [workflow] table.{doc_workspace} is the run folder for this workflow.Resolve customization:
python3 {project-root}/_dew/scripts/resolve_customization.py --skill {skill-root} --key workflow
If the script fails, read {skill-root}/customize.toml directly and use defaults.
Execute {workflow.activation_steps_prepend}.
Load persistent facts from {workflow.persistent_facts}.
Load {project-root}/_dew/dew/config.yaml if present.
Load:
{workflow.serving_spec_template}{workflow.metric_contract_template}{workflow.trust_limitations_template}{workflow.serving_mode_rubric}{workflow.self_service_rubric}Greet user in configured language.
Read fully and follow:
steps/step-01-init.md