Name: databricks-dbsql
Rating: 68 (1 reviews)
Author: databricks

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.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

80%Weight 40%Scale 1-5

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

The body is a highly actionable, well-structured overview with executable examples for every feature and a clean one-level-deep reference structure (all referenced files exist). Its weakness is workflow clarity: it is a recipe catalog without explicit validation/feedback checkpoints for its batch and destructive operations.

Suggestions

Add a short 'Develop & deploy workflow' section with explicit checkpoints (write SQL -> test via `databricks experimental aitools tools query` with LIMIT -> validate row counts/schema -> promote), especially before destructive MERGE/scheduled-refresh operations.

Trim the longest inline recipes (geospatial proximity join, full stored procedure) to minimal illustrative snippets and point to the reference files for the complete versions, to lift conciseness toward the lean anchor.

Dimension	Reasoning	Score
Conciseness	The body is well-organized with a compact Quick Reference table and lean Key Guidelines, and it assumes Claude knows SQL rather than explaining it; a few of the longer inline code recipes (geospatial join, full stored procedure) could be trimmed or moved to references, so it sits just below the lean anchor.	4 / 5
Actionability	Provides copy-paste-ready, executable SQL covering the common cases (scripting, stored procedures with error handling, materialized views with scheduling, pipe syntax, AI functions, geospatial, collation, http_request, read_files, recursive CTE, remote_query), matching the fully-executable anchor.	5 / 5
Workflow Clarity	This is a feature catalog of self-contained recipes rather than a sequenced multi-step workflow, and while the stored-procedure example includes an EXIT HANDLER and Key Guidelines mention testing via CLI before deploying, there are no explicit validate-fix-retry checkpoints for the batch/destructive operations (MERGE upsert, scheduled MV refresh), which caps this at the anchor-3 level.	3 / 5
Progressive Disclosure	Clear overview (Quick Reference table) with well-signaled one-level-deep references, and the Reference Files table maps each of the five real, verified reference files to its contents and 'When to Read' guidance, matching the clear-overview anchor.	5 / 5
	Total	17 / 20 Passed

Description

86%Weight 40%Scale 1-5

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

The description is a strong, trigger-rich statement of when to invoke the skill, with comprehensive natural-language keywords and synonyms and an explicit what/when structure. Its main weakness is that capabilities are listed as trigger terms rather than concrete actions, keeping specificity at the midpoint.

Suggestions

Lead the description with 1-2 concrete verbs describing what the skill does (e.g., 'Write, optimize, and deploy advanced Databricks SQL...') before the trigger list, to raise specificity beyond a feature catalog.

Tighten the 'SHOULD also invoke' clause to advanced-SQL-specific contexts to reduce overlap with the parent databricks-core skill.

Dimension	Reasoning	Score
Specificity	Names the domain ('Databricks SQL (DBSQL) advanced features and SQL warehouse capabilities') and enumerates concrete features (materialized view, stored procedure, geospatial, AI functions), but these surface as trigger terms rather than described actions, so it sits between anchor 2 and 3.	3 / 5
Completeness	Explicitly answers both 'what' (DBSQL advanced features and SQL warehouse capabilities) and 'when' ('MUST be invoked when the user mentions...' plus 'SHOULD also invoke when...'), matching the anchor for clear what-and-when with concrete trigger phrases.	5 / 5
Trigger Term Quality	Comprehensive list of natural terms with synonyms and variations (e.g., 'temp table'/'temporary view', 'recursive CTE'/'WITH RECURSIVE', 'materialized view'/'CREATE MATERIALIZED VIEW', 'pipe syntax'/'\|>'/'pipe operator') covering exactly what a user would say.	5 / 5
Distinctiveness Conflict Risk	Clear niche (Databricks SQL advanced features) with specific triggers and minimal conflict risk, but the broader 'SHOULD also invoke when the user asks about SQL best practices, data modeling patterns, or advanced SQL features on Databricks' clause creates minor overlap risk with a parent databricks-core skill.	4 / 5
	Total	17 / 20 Passed

Validation

93%

Warnings & errors only

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

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: databricks/devhub
Path: .databricks/aitools/skills/databricks-dbsql/SKILL.md
Commit: ade58e4

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

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.