A production-minded data quality workspace for completeness, uniqueness, validity, freshness, drift, run history, and local CSV profiling.
- Dataset Inventory: Searches data products by name, owner, and source.
- Quality Dimensions: Separates completeness, uniqueness, validity, freshness, and drift.
- Rule Evidence: Shows threshold score, failure count, column, and status.
- Run History: Adds deterministic validation runs with duration and row context.
- Trend Visualization: Charts composite quality score over recent runs.
- CSV Profiler: Calculates completeness and row uniqueness from local files.
- Operational Summary: Aggregates datasets, scores, failing rules, and observed rows.
- Validated Persistence: Stores compatible dataset state in IndexedDB.
public/favicon.svg
public/sample-customer-import.csv
public/sample-quality-datasets.json
src/app.tsx
src/components/dataset-list.tsx
src/components/quality-panel.tsx
src/domain/quality.ts
src/index.css
src/main.tsx
src/services/quality-store.ts
- Preact 10 with Vite 8
- TypeScript 6
- Tailwind CSS 4
- Prettier 3
- simple-statistics for dimension summaries
- Chart.js for quality-run history
- Papa Parse, Zod, IndexedDB, nanoid, date-fns, and Lucide icons
public/sample-quality-datasets.json includes three synthetic warehouse datasets with healthy, warning, and failing rules. public/sample-customer-import.csv contains missing and duplicate rows for profiling.
All organizations, identities, amounts, events, and operational records are synthetic and remain in the browser.
npm install
npm run devOpen http://localhost:5173.
- Replace
src/services/quality-store.tswith orchestration, warehouse, and observability APIs. - Imported CSV files remain in the browser and only the first rows are retained as sample context.
- Production checks should preserve query, rule version, partition, lineage, and reproducible run evidence.
- Treat quality scores as decision support; critical business rules should keep explicit pass/fail semantics.
- Profile large files in a worker or streaming engine rather than loading them fully into memory.
- Dense rule tables scroll horizontally on mobile while the dataset navigator stays usable.
npm run format:check
npm run build
npm run preview
npm audit --omit=dev{
"$schema": "https://openapi.vercel.sh/vercel.json",
"framework": "vite",
"buildCommand": "npm run build",
"outputDirectory": "dist",
"rewrites": [
{
"source": "/(.*)",
"destination": "/index.html"
}
]
}MIT License. See LICENSE.