Examples & Use Cases
End-to-end scenarios that show the Qualytics CLI in action: real customer workflows, basic to advanced. Pick the scenario that matches what you're trying to do; each one is a self-contained guide with the full CLI sequence, the API endpoints called behind the scenes, a Python equivalent, the minimum permissions required, and troubleshooting for the most common failures.
How each page is structured
Every example below follows the same shape: Goal → Permissions → Prerequisites → CLI workflow → Behind the scenes → Python equivalent → Variations → Troubleshooting → Related. Skim the section that matters most for your situation; the rest is there when you need it.
Datastores
Onboard new sources, one at a time or in bulk.
| Scenario | What it covers |
|---|---|
| Onboard a single datastore | Connection → datastore → sync → profile → scan, plus DFS and IAM-role variations. |
| Bulk datastore onboarding | Define many datastores in YAML, apply the whole folder with config import. |
Operations
Run the recurring data quality pipeline.
| Scenario | What it covers |
|---|---|
| Daily sync, profile, and scan | The standard three-step pipeline, foreground and background, single and multi-datastore. |
| Targeted scans | Scan only specific containers or tag-filtered subsets. |
| Incremental scans for large tables | --incremental with --greater-than-time or --greater-than-batch for billion-row tables. |
Quality Checks
Author, promote, and govern checks.
| Scenario | What it covers |
|---|---|
| Bulk-create quality checks | Create many checks in one shot from a YAML library. |
| Promote checks Dev to Prod | Export, version in Git, import to Prod with idempotent upsert. |
| Audit and clean up draft checks | Review the queue of Draft checks, promote keepers, archive the rest. |
| Manage check templates | Export, import, and govern reusable check templates. |
Anomalies
Triage at scale.
| Scenario | What it covers |
|---|---|
| Bulk anomaly triage | Filter, bulk update with assignees, bulk archive with the right outcome status. |
| Daily triage automation | Cron-driven script that auto-archives stale anomalies and routes fresh ones to on-call. |
Computed Containers
SQL-defined virtual tables and joins.
| Scenario | What it covers |
|---|---|
| Bulk import computed tables from CSV | Drop a CSV of name/description/query rows; the CLI registers them all and creates checks per row. |
| Build a computed join | Combine two existing containers into a new joined view without writing the SQL by hand. |
Configuration as Code
Treat the whole Qualytics config as version-controlled YAML.
| Scenario | What it covers |
|---|---|
| Export and import full configuration | Connections, datastores, containers, fields, and checks exported as a Git-friendly folder, importable into any environment. |
| Drift detection between environments | Detect when the live config has been changed in the UI without going through Git. |
Automation
Schedule recurring work and integrate with CI.
| Scenario | What it covers |
|---|---|
| Scheduled metadata exports | Daily/hourly snapshots of anomalies, checks, and field profiles to your enrichment datastore. |
| GitHub Actions pipelines | PR validation, tagged-release promotion to Prod, nightly scans, drift checks. |
AI Integration
Let Claude Code, Cursor, and other AI assistants use Qualytics directly.
| Scenario | What it covers |
|---|---|
| Connect MCP clients | Wire qualytics mcp serve into Claude Code or any MCP-aware client. |
Permission and role primer
Most CLI commands check two kinds of permissions before they do anything. Each example page lists the minimums for the endpoints it calls.
| Layer | What it controls | Levels |
|---|---|---|
| User role (global) | What kinds of resources you can touch at all | Admin > Manager > Member |
| Team permission (per resource) | What you can do with a specific resource based on your team membership | Editor > Author > Drafter > Viewer > Reporter |
If you're not sure what role your token has, run:
For the full conceptual model see Team Permissions.