Add databricks-dqx skill for Data Quality Extensions

[aniljoshiaset]

Summary

• Adds comprehensive skill for Databricks Labs DQX (Data Quality Extensions) framework
• Covers quality rule definition, profiling, auto-generation, Lakeflow/DLT integration, streaming, and check functions
• SKILL.md (256 lines) with 7 progressive-disclosure reference files following skill authoring best practices

Skill Structure

databricks-skills/databricks-dqx/
├── SKILL.md                          # Main skill (256 lines)
├── 1-installation-setup.md           # Install methods (pip, CLI, DABs, extras)
├── 2-defining-quality-rules.md       # DQRule classes, YAML metadata, storage
├── 3-applying-checks.md              # Apply checks, split valid/invalid, save
├── 4-profiler-auto-generation.md     # Profile datasets, auto-generate rules
├── 5-lakeflow-integration.md         # DLT/Lakeflow pipeline integration
├── 6-streaming-metrics.md            # Streaming support, quality monitoring
└── 7-check-functions-reference.md    # Complete built-in check function catalog

Test plan

• SKILL.md has valid YAML frontmatter with name and description
• Description is third-person, includes what the skill does and when to use it
• Under 500 lines (256)
• Progressive disclosure with reference files one level deep
• No time-sensitive information
• All file paths use forward slashes
• Updated databricks-skills/README.md skills table
• Updated main README.md skill count (19 → 20)

🤖 Generated with Claude Code

[calreynolds]

@aniljoshiaset instead of having us build in dqx into ai-dev-kit, would it be possible to create a dqx skill \ set of skills that we can reference on install from your repo as a part of our install.sh? Want to make sure we reduce areas of maintenance as much as possible 👍

[aniljoshiaset]

@calreynolds Absolutely! I've set up a standalone repo for the DQX skill here: https://github.com/aniljoshiaset/databricks-dqx-skill

The structure follows the same pattern as the MLflow skills — a databricks-dqx/ directory with SKILL.md + 7 reference guides that can be fetched via raw GitHub URLs:

databricks-dqx/
├── SKILL.md                        # Main orchestrator
├── 1-installation-setup.md
├── 2-defining-quality-rules.md
├── 3-applying-checks.md
├── 4-profiler-auto-generation.md
├── 5-lakeflow-integration.md
├── 6-streaming-metrics.md
└── 7-check-functions-reference.md

You should be able to add it to install.sh the same way you reference MLflow skills. Let me know if you need any changes to the structure!