databricks-dbsql
Databricks SQL (DBSQL) advanced features and SQL warehouse capabilities. This skill MUST be invoked when the user mentions: "DBSQL", "Databricks SQL", "SQL warehouse", "SQL scripting", "stored procedure", "CALL procedure", "materialized view", "CREATE MATERIALIZED VIEW", "pipe syntax", "|>", "geospatial", "H3", "ST_", "spatial SQL", "collation", "COLLATE", "ai_query", "ai_classify", "ai_extract", "ai_gen", "AI function", "http_request", "remote_query", "read_files", "Lakehouse Federation", "recursive CTE", "WITH RECURSIVE", "multi-statement transaction", "temp table", "temporary view", "pipe operator". SHOULD also invoke when the user asks about SQL best practices, data modeling patterns, or advanced SQL features on Databricks.
85
81%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description excels at trigger term coverage and explicit invocation guidance, making it very effective for skill selection from a large pool. Its main weakness is that the 'what does this do' portion is vague—it says 'advanced features and SQL warehouse capabilities' without listing concrete actions like 'creates materialized views, executes stored procedures, runs geospatial queries.' The description is heavily weighted toward trigger terms at the expense of describing actual capabilities.
Suggestions
Replace the vague opening 'advanced features and SQL warehouse capabilities' with specific concrete actions, e.g., 'Creates and manages materialized views, executes stored procedures, performs geospatial queries with H3/ST_ functions, calls AI functions, and handles multi-statement transactions in Databricks SQL.'
Consider grouping the trigger terms into logical categories (e.g., 'SQL scripting: stored procedures, CALL procedure; Geospatial: H3, ST_, spatial SQL') to improve readability while maintaining the comprehensive coverage.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain (Databricks SQL) and mentions 'advanced features and SQL warehouse capabilities,' but the actual concrete actions are not listed—it focuses on trigger terms rather than describing what the skill does (e.g., 'creates materialized views,' 'executes stored procedures'). The capabilities are implied through keyword lists rather than explicitly stated as actions. | 2 / 3 |
Completeness | The description answers both 'what' (Databricks SQL advanced features and SQL warehouse capabilities) and 'when' with an extensive explicit trigger list using 'MUST be invoked when' and 'SHOULD also invoke when' clauses. The 'what' is somewhat thin but the 'when' is exceptionally detailed. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would actually say, including abbreviations ('DBSQL'), full names ('Databricks SQL'), specific SQL syntax ('|>', 'WITH RECURSIVE', 'ST_'), function names ('ai_query', 'ai_classify'), and conceptual terms ('Lakehouse Federation', 'geospatial'). Very comprehensive keyword coverage. | 3 / 3 |
Distinctiveness Conflict Risk | The description is highly distinctive with very specific Databricks SQL terminology and function names that are unlikely to conflict with generic SQL skills or other tools. The trigger terms like 'DBSQL', 'ai_query', 'Lakehouse Federation', and 'pipe syntax |>' create a clear, unique niche. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
72%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured reference skill with excellent progressive disclosure and highly actionable code examples covering a broad range of DBSQL features. Its main weakness is that the inline examples section is quite long (potentially better to trim a few and rely more on the reference files), and multi-step workflows lack explicit validation checkpoints. The Key Guidelines section provides useful guardrails but could be better integrated into the workflow examples.
Suggestions
Add validation/verification steps to multi-step examples (e.g., after CREATE PROCEDURE, show a test CALL with expected output; after CREATE MATERIALIZED VIEW, show how to verify refresh status)
Consider trimming 2-3 of the simpler inline examples (e.g., consolidate read_files JSON and CSV into one example, or move to the reference file) to improve conciseness while relying on the well-structured reference files for details
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient with good use of tables and code examples, but the sheer volume of examples (13 distinct patterns) makes it lengthy. Some examples like read_files CSV and JSON could be consolidated. However, it avoids explaining basic concepts Claude already knows. | 2 / 3 |
Actionability | Every feature includes fully executable, copy-paste-ready SQL examples with realistic table references, proper syntax, and complete patterns including error handling, invocation syntax, and connection setup. The examples are concrete and specific. | 3 / 3 |
Workflow Clarity | Individual examples are clear but multi-step workflows lack explicit validation checkpoints. For instance, the stored procedure example doesn't include a test/validate step after creation, and the http_request pattern doesn't mention verifying the connection setup. The Key Guidelines mention using MCP tools to validate but this isn't woven into the workflows themselves. | 2 / 3 |
Progressive Disclosure | Excellent progressive disclosure with a clear quick reference table at top, common patterns inline for immediate use, and a well-organized reference files table with specific 'When to Read' guidance pointing to one-level-deep detail files. Navigation is intuitive and clearly signaled. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- databricks-solutions/ai-dev-kit
- Commit
02aac8c
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.