CSAV-DP

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.

Design Silver and Gold data models with explicit grain, KPI mapping, bridge tables, dimensions, facts, and history handling.

Create, update, or validate a data product PRD with consumers, KPI hypotheses, grain, freshness, trust expectation, known limitations, and evidence requirements.

Define data quality rules, severity, actions, evidence requirements, and test handoff for source, transformation, and serving layers.

Design operational readiness, monitoring, alerting, incident handling, backfill, change process, environments, and CI/CD handoff.

Run a mandatory DEW decision gate. Use when a project-shaping data engineering decision must be explained, optioned, approved by the user, and recorded before continuing.

Implement a ready-for-dev data engineering story while producing required evidence, logs, tests, and story updates.

Validate the implemented data product from source through storage, ingestion, transformation, DQ, Gold, and serving outputs.

Check whether a DEW phase has the validation evidence required before it can be marked complete.

Create a quickstart example project walkthrough showing how to use DEW from brief to release.

Review whether all required implementation, validation, DQ, serving, governance, and operational evidence exists before release.

Design privacy posture, access control, ownership, lineage, metadata, auditability, and governance rules for the data product.

Resolve a DEW HALT by explaining the blocker, options, evidence requirements, and resume conditions.

Analyzes current DEW state and user query to recommend the next skill(s) or answer workflow questions.

Review whether generated stories have enough context, evidence criteria, dependencies, and DoD before development starts.

Generates or updates an index.md to reference all docs in a target folder.

Design ingestion patterns, source extraction, frequency, schema evolution, failure handling, idempotency, replay, and evidence requirements.

Create installation, usage, command reference, and troubleshooting documentation for the DEW skill module.

Define candidate KPIs, formulas, grains, required fields, and source dependencies without claiming feasibility.

Validate whether candidate KPIs can be computed from available fields, samples, and source evidence before architecture.

Create executable SQL, Python, or notebook prototype to compute KPI on sample data and produce validation evidence.

Review KPI feasibility evidence, assign feasibility status, and trigger HALT if evidence is missing.

Capture learning retrospective, FDE concept mapping, mistakes, decisions, and skill growth from the completed project.

Package the DEW skill module for release with versioning, release notes, checklist, and package readiness gate.

Extract reusable data engineering patterns, anti-patterns, and templates from the completed DEW project.

Create, update, or validate a DEW project brief focused on data engineering scope, business purpose, learning objective, and evidence-driven continuation.

Create user-facing and technical documentation for the released data product from validated artifacts, evidence, and release notes.

Decide whether the data product is ready to release with scope, caveats, risks, notes, and rollback guidance.

Review KPI feasibility, source validation, trust, grain, freshness, caveats, and MVP scope before allowing architecture.

Perform a critical adversarial review and produce a findings report.

Walk branching paths and boundary conditions in content, reporting only unhandled edge cases.

Design how Gold models, metrics, and data products are served to consumers with trust, caveats, contracts, and feedback loop.

Splits large markdown documents into smaller organized files based on heading sections.

Audit DEW skill pack structure, module catalog, cross-links, gates, HALTs, assets, steps, and checklist consistency before release.

Run or document source access checks such as API call, file download, database query, or manual acquisition.