datacoolie

Build and maintain an internal technical LLM wiki when the user requests wiki work or a verified major or architectural change qualifies for automatic wiki review. Use for project memory, architecture notes, contracts, runbooks, decisions, glossary, incremental ingest, delta status, and wiki health. Do not use merely because a routine task or plan completed.

Explore ambiguous ideas and solution directions before evidence-backed research or implementation planning. Use for ideation, framing, assumption checks, option generation, product or architecture trade-offs, and deciding whether research or a plan is needed.

Research technical options with source-backed recommendations. Use for technology evaluation, architecture trade-offs, library/framework choices, best practices, scalability, security, maintainability, or when current external facts matter.

Create implementation plans, architecture decisions, and phased roadmaps. Use for Standard or Complex changes, new pipelines, schema changes, migrations, multi-file features, technology choices, or when acceptance criteria and verification strategy are unclear.

Review code and data pipeline changes for correctness, security, performance, maintainability, contract breaks, and missing verification. Use before merge, after implementation, for PRs, commits, pending diffs, or codebase risk scans.

Design data ingestion into landing/bronze layers. Use for source onboarding, transfer pattern selection, full/incremental/CDC scope, change detection, landing zones, schema drift, idempotent file or batch processing, and ingestion quality gates.

Design warehouse and semantic data models. Use for grain definition, fact/dimension design, star schema, Kimball, Data Vault, SCD, bridge tables, surrogate keys, semantic metrics, and model review.

Define and enforce data quality rules, contracts, assertions, reconciliation, quarantine, and quality gates. Use for completeness, uniqueness, validity, freshness, schema, referential integrity, SCD checks, or source-to-target validation.

Create and modify Power BI report files in PBIR/PBIP format using the `powerbi-report-author` and `powerbi-desktop` CLIs. Use when the user wants to: (1) implement an approved report spec or design brief, (2) add or edit pages, visuals, filters, slicers, bookmarks, themes, or formatting, (3) validate PBIR and verify rendering in Power BI Desktop. For open-ended visual design, use `power-bi-report-design` first. For end-to-end requirements and approval workflow, use `powerbi-report-planning` first. Triggers: "edit PBIR", "create Power BI report page", "add visual to PBIP", "format report visual", "validate Power BI report", "reload Desktop screenshot", "implement an approved PBIP report spec", "edit PBIR pages/visuals".

Analyze business requirements and define BI strategy for Power BI projects. Use this skill whenever the user wants to analyze business requirements, identify KPIs, define what insights a report should deliver, propose dashboard structure for a business domain, or gather requirements before building a Power BI report or semantic model. Triggers include: "analyze business requirements", "what KPIs should we track", "what insights can we get", "business domain analysis", "requirements gathering", "what should we measure", "what problems can we solve", "propose a dashboard", "what reports do we need", "BI strategy", "define metrics". Do NOT use for actual model building (use power-bi-semantic-model), DAX writing (use power-bi-dax-development), or report generation (use power-bi-report-authoring).

Reference kit for the feedback and iteration phase of a Power BI project. Provides classification taxonomy, prioritization matrix, intake template, change-impact scoping, post-change validation checklist, A/B variant testing workflow, formal UAT workflow, PBIP Git diff guide, changelog template, and a performance quick-check pointer. Use this skill whenever the user provides feedback on an existing Power BI report, requests iteration, wants to run UAT, or needs to document changes. Routing (feedback → correct downstream skill) is owned by the power-bi-developer agent, not by this skill. Triggers include: "user feedback", "improve report", "fix report", "iterate on the report", "report review", "rebuild report", "UAT", "changelog", "release notes", "users are complaining", "performance issue", "add/remove/modify a page/measure/visual on an existing report". Do NOT use this skill for brand-new projects (use power-bi-business-analysis to start from scratch).

Diagnose and resolve Power BI performance issues across models, DAX queries, report visuals, and data refresh. Use this skill whenever the user reports slow report loading, slow visual interactions, long refresh times, high memory usage, query timeouts, or capacity bottlenecks. Triggers include: "report is slow", "performance issue", "optimize report", "slow loading", "query timeout", "refresh takes too long", "high memory", "report performance", "visual is slow", "optimize model", "reduce model size", "aggregation table", "incremental refresh", "Performance Analyzer", "DAX Studio", "server timings", "capacity metrics". Do NOT use for initial model design (use power-bi-semantic-model), initial DAX development (use power-bi-dax-development), or report design decisions (use power-bi-report-design). This skill is for diagnosing and fixing performance problems in existing solutions.

Design Power BI report layouts, select chart types, plan page structures, choose themes, and apply data storytelling principles BEFORE generating PBIR JSON files. Use this skill whenever the user asks to design a report, plan a dashboard layout, choose visualizations, decide on chart types, apply storytelling to data, select a theme or color palette, plan page navigation, or structure report pages for a Power BI project. Also use when the user provides requirements and needs a report design spec before generation, or when reviewing/improving an existing report's visual design, layout, or UX. This skill produces a Design Spec — a structured plan of pages, visuals, layout, theme, and navigation — that feeds into the `power-bi-report-authoring` skill for JSON generation. Do NOT use for generating PBIR JSON files, validating report structure, or building the semantic model. For JSON generation, use `power-bi-report-authoring`.

Design and build Power BI semantic models using star schema principles and the PowerBI Modeling MCP tools. Use this skill whenever the user wants to build a semantic model, create a data model, design a star schema, add tables or columns, create or modify relationships, configure storage modes (Import, DirectQuery, DirectLake, Composite), implement RLS (Row-Level Security), optimize model performance, or explore existing data sources for modeling. Triggers include: "build semantic model", "create data model", "star schema", "add table", "create relationship", "storage mode", "DirectLake", "composite model", "RLS", "optimize model", "model review", "explore data", "connect to gold layer", "extend model", "add dimension", "add fact table". Do NOT use for DAX measure creation (use power-bi-dax-development) or report generation (use power-bi-report-authoring).

Develop, optimize, and validate DAX measures, calculation groups, visual calculations, field parameters, dynamic format strings, time intelligence, semi-additive logic, virtual relationships with TREATAS, DAX window functions, and user-defined functions for Power BI semantic models. Use for requests to write or optimize DAX, create measures, explain CALCULATE or evaluation context, build YTD/YoY/WTD logic, use RUNNINGSUM or MOVINGAVERAGE, rank with RANK/ROWNUMBER/OFFSET/INDEX/WINDOW, or apply advanced DAX patterns. Research Microsoft Learn MCP before recommending patterns.

Generate, validate, convert, and merge DataCoolie ETL metadata files. Use when user asks to create, build, author, or edit metadata (connections, dataflows), validate metadata against JSON Schema, convert JSON↔YAML↔Excel, merge environment overlays, or refresh schema from GitHub with --fetch-latest.

Design data platform architecture based on discovery findings. Supports incremental source onboarding — add new sources without redoing approved decisions. Reads discovery reports and produces a medallion architecture design document with stage-level contracts, engine strategy, infrastructure requirements, and partitioning. Use when user says "design architecture", "medallion", "plan layers", "add source", "what infrastructure do I need", "how should I structure the pipeline", "bronze silver gold", "architecture design", "update architecture", or after completing source discovery.

Deploy datacoolie ETL jobs to AWS Glue, Databricks, and Microsoft Fabric. AI-driven workflows: preflight checks, runner generation, functions packaging, artifact upload, job creation/update, promotion, and CI/CD workflow generation. Use when user says "deploy", "ship to prod", "apply", "preflight", "generate runner", "package functions", "cicd", "aws glue deploy", "fabric deploy", "databricks bundle", or mentions shipping datacoolie to any cloud platform.

Discover and profile data sources for DataCoolie ETL projects. Produces per-source discovery reports (operational context + schema inventory). Two retrieval methods: auto-introspection (connect to source) or interview (ask user). Supports multiple sources — each source gets its own report. Use when user says "discover sources", "explore data", "profile source", "what tables exist", "understand my data", "source assessment", "new data source", "connect to source", or before designing architecture for a new data platform.

Bootstrap DataCoolie ETL project workspace files with best-practice structure. Consumes the approved architecture document to pre-populate config and runner source, environment overlays only when those artifacts are needed. Falls back to interactive mode when no architecture exists. Use when user says "scaffold project", "init", "initialize", "create project", "setup datacoolie", "new project", or "start new ETL".

Provision platform infrastructure from an approved architecture design. Creates lakehouses, warehouses, S3 buckets, databases, or local directories. Supports CLI mode (execute commands) and Terraform mode (generate .tf files). Use when user says "provision", "create resources", "setup infrastructure", "create lakehouse", "create bucket", "terraform", "infra", "provision resources", or after architecture approval.