CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

databricks-unity-catalog

Unity Catalog governance, access control, and observability. Use to grant or revoke access (GRANT/REVOKE), reason about the privilege model and ownership, set up row-level security and column masks, create external locations and storage credentials, define catalogs/schemas/tables/volumes, answer "who can read this table", query system tables (audit, lineage, billing), work with volume files in /Volumes/, and set up connection passwords (secret scopes, Azure Key Vault, AWS Secrets Manager).

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

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.

A strong reference-style skill body: highly actionable commands and queries, excellent progressive disclosure across 8 real reference files, and lean writing throughout. The main gap is workflow validation — governance changes (GRANT/REVOKE) lack an explicit verify-after-change step — plus some repeated version-caveat text.

Suggestions

Add an explicit verification step after GRANT/REVOKE examples (e.g., re-run `SHOW GRANTS ON …` and confirm the principal's privileges changed as expected) so destructive governance operations have a validate→fix loop.

Consolidate the three repeated "CLI surface varies by version" warnings into the single opening note and reference it from the other spots, trimming redundant tokens.

State the boundary with databricks-core in the When to Use section (exploration vs. governance) to reduce trigger ambiguity between the two skills.

DimensionReasoningScore

Conciseness

The body is dense and assumes Claude's competence — no re-teaching of concepts, tight commented code blocks — but the "CLI surface varies by version" caveat is repeated three times (opening note, note after the positional-args table, and the end of the Running SQL section), which is a minor trim candidate. This matches the efficient-with-minor-overexplanation anchor rather than the every-token-earns-its-place anchor.

4 / 5

Actionability

Guidance is fully executable and copy-paste ready: version-stable `--json` create commands, `fs` operations with the required `dbfs:/Volumes/` prefix, complete GRANT/REVOKE/SHOW GRANTS SQL, runnable Python SDK fallback, and even how to obtain WAREHOUSE_ID/METASTORE_ID. The examples cover the common cases for each sub-topic.

5 / 5

Workflow Clarity

The body includes a pre-flight CLI version check and fallback-on-failure guidance, but for the core GRANT/REVOKE governance operation the validation checkpoint is only implicit — SHOW GRANTS appears as a demo query in the same code block rather than as an explicit post-change verification step. Since REVOKE is a destructive database operation, the rubric's cap on missing validation in such workflows applies; this fits the "sequence present but checkpoints implicit" anchor rather than the explicit-verification anchor.

3 / 5

Progressive Disclosure

All 8 referenced files exist in references/ and are one level deep, clearly signaled both in a well-organized topic/description table and via inline links where relevant (e.g., Best Practice 6 pointing at references/8-connections-and-secrets.md). SKILL.md stays an overview with quick-start examples while bulk detail lives in the reference files, matching the clear-overview anchor.

5 / 5

Total

17

/

20

Passed

Description

96%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.

An excellent description: concrete, comprehensive, and trigger-rich, with an explicit "Use to…" clause and a natural quoted user question. The only weakness is undelineated overlap with the databricks-core parent around catalog/schema/table creation and exploration.

DimensionReasoningScore

Specificity

The description lists many concrete actions — "grant or revoke access (GRANT/REVOKE)", "row-level security and column masks", "create external locations and storage credentials", "define catalogs/schemas/tables/volumes", "query system tables (audit, lineage, billing)" — giving comprehensive coverage of the domain, matching the anchor for multiple specific concrete actions rather than leaving only minor gaps.

5 / 5

Completeness

It explicitly answers "what" ("Unity Catalog governance, access control, and observability") and "when" via an explicit "Use to…" clause with concrete trigger phrases, matching the anchor that clearly and explicitly answers both with concrete triggers.

5 / 5

Trigger Term Quality

Natural user phrasing is comprehensively covered, including the quoted question "who can read this table", "row-level security", "column masks", "audit", "lineage", "billing", and synonyms for secret backends ("secret scopes, Azure Key Vault, AWS Secrets Manager"). It goes beyond the good-but-incomplete keyword coverage of the level-4 anchor.

5 / 5

Distinctiveness Conflict Risk

The Unity Catalog governance niche is clear with distinct triggers, but "define catalogs/schemas/tables/volumes" overlaps with the declared parent skill databricks-core (which owns catalog/table exploration per the body), and the description does not delineate that sibling boundary — minor overlap risk with a closely related skill rather than minimal conflict risk.

4 / 5

Total

19

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 11 suspicious

Warning

Total

14

/

16

Passed

Repository

databricks/databricks-agent-skills

Path

plugins/databricks/claude/skills/databricks-unity-catalog/SKILL.md

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.