CtrlK
BlogDocsLog inGet started
Tessl Logo

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.

64

Quality

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Advisory

Suggest reviewing before use

SKILL.md
Quality
Evals
Security

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable, packed with executable SQL, but its reference structure is undercut by missing bundle files and it lacks an explicit validation feedback loop for the destructive/batch SQL operations it demonstrates.

Suggestions

Consolidate the Quick Reference and Reference Files tables, which both list the same five files and their contents, into a single table to remove redundancy and tighten the SKILL.md body.

Add an explicit validation feedback loop for destructive/batch SQL (e.g. MERGE, INSERT, CREATE PROCEDURE): a sequenced 'run via execute_sql_multi -> review errors -> fix -> re-validate' workflow rather than a single guideline bullet.

Add the five referenced files (sql-scripting.md, materialized-views-pipes.md, geospatial-collations.md, ai-functions.md, best-practices.md) to the bundle so the progressive disclosure chain resolves, and consider moving some of the inline Common Patterns examples into those files to slim the overview.

DimensionReasoningScore

Conciseness

The body is mostly dense, executable SQL with little conceptual padding, but the Quick Reference and Reference Files tables redundantly enumerate the same five files, and ~11 full inline code examples overlap the reference files' purpose, so it could be tightened rather than earning the lean anchor 3.

2 / 3

Actionability

It provides fully executable, copy-paste-ready SQL across stored procedures (with DECLARE EXIT HANDLER), materialized views with scheduling, pipe syntax, AI functions, http_request, read_files, and recursive CTEs, matching the executable anchor 3.

3 / 3

Workflow Clarity

It is organized as a pattern catalog rather than a sequenced workflow, and validation appears only as a single guideline bullet ('Use MCP tools (execute_sql, execute_sql_multi) to test and validate all SQL before deploying') with no structured validate->fix->retry loop for destructive/batch ops like MERGE and INSERT, capping it at 2.

2 / 3

Progressive Disclosure

The structure is well-signaled with a one-level-deep 'When to Read' reference table (an anchor-3 pattern), but verification shows the five referenced files (sql-scripting.md, materialized-views-pipes.md, etc.) are not present in the bundle, so the disclosure chain does not resolve to real content.

2 / 3

Total

9

/

12

Passed

Description

90%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with explicit, well-scoped trigger guidance and broad natural keyword coverage. Its only weakness is that the 'what it does' sentence uses abstract feature-naming rather than concrete action verbs.

DimensionReasoningScore

Specificity

The opening 'Databricks SQL (DBSQL) advanced features and SQL warehouse capabilities' names the domain and many feature areas but uses abstract capability-naming rather than concrete action verbs (e.g. 'create stored procedures', 'build materialized views'), so it sits above the vague anchor 1 but below the action-listing anchor 3.

2 / 3

Completeness

It explicitly answers both what (advanced features and SQL warehouse capabilities) and when ('This skill MUST be invoked when the user mentions:...', 'SHOULD also invoke when...'), with explicit trigger guidance that is not capped at 2.

3 / 3

Trigger Term Quality

It provides broad, natural keyword coverage users would actually say ('DBSQL', 'Databricks SQL', 'stored procedure', 'materialized view', 'geospatial', 'H3', 'recursive CTE', 'WITH RECURSIVE'), matching the good-coverage anchor 3.

3 / 3

Distinctiveness Conflict Risk

It occupies a clear Databricks-specific niche with distinctive triggers (H3, '|>', ai_query, COLLATE) unlikely to fire for unrelated skills; a few generic terms like 'temp table' could overlap a general SQL skill but the Databricks-specific terms dominate.

3 / 3

Total

11

/

12

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 20 missing

Warning

Total

15

/

16

Passed

Repository
databricks-solutions/ai-dev-kit
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.