CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

databricks-lakeflow-connect

Lakeflow Connect managed ingestion into Unity Catalog from databases (SQL Server, PostgreSQL, MySQL, Oracle) by CDC or query-based. Use when creating, running or debugging these pipelines.

Quality

91%

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

92%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.

An exceptionally well-engineered skill body: dense, non-obvious guidance with explicit gates and validation checkpoints, backed by a real, correctly-anchored one-level-deep reference bundle. The only body-level weakness is that complete end-to-end examples live in references rather than inline.

DimensionReasoningScore

Conciseness

Lean and dense throughout: 'The docs pages win' frames precedence in one line, tables carry only non-obvious facts (marking fields, per-source pages, mode gates), and no concept Claude already knows is explained. Every token earns its place.

5 / 5

Actionability

Mostly executable: concrete commands ('databricks bundle run <key> --no-wait', 'pipelines get-update <id> <update-id> -o json', 'ALTER CONNECTION <name> OPTIONS (..., password secret(\'<scope>\', \'<key>\'))') and exact YAML field names ('connector_type: CDC', 'ingestion_gateway_id: ${resources.pipelines.<gateway>.id}'). Not 5 because complete end-to-end examples are delegated to reference files and external docs pages rather than shown in the body — a minor gap when scoring the body alone.

4 / 5

Workflow Clarity

Clear sequence (tool selection → mode choice → compute → 6-item pre-creation checklist gated on one 'yes' → run/verify → failure decision tree with 'Fix that one thing and retry the same step'), with explicit validation checkpoints (row-count comparison after loads, post-update compute confirmation, query-based pre-flight checks) and confirmation gates for destructive operations (full refresh warning, cleanup with user OK, slot SQL handed to the DBA).

5 / 5

Progressive Disclosure

The ~90-line body is an overview with a 'Reference files' table (file + when-to-use) plus inline anchored links at point of use; all four references (modes.md, sources.md, troubleshooting.md, rows-per-run.md) and scripts/rows_per_run.sql exist, and every linked anchor resolves. References are one level deep with only lateral cross-links, matching the bundle structure.

5 / 5

Total

19

/

20

Passed

Description

87%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.

A strong description: it states what the skill does with specific sources and modes, and gives an explicit 'Use when creating, running or debugging these pipelines' trigger clause. Only minor improvements possible via a few extra natural synonyms.

DimensionReasoningScore

Specificity

Lists several concrete actions — 'managed ingestion into Unity Catalog from databases (SQL Server, PostgreSQL, MySQL, Oracle) by CDC or query-based' plus 'creating, running or debugging these pipelines' — with the source databases enumerated. Not 5 because it omits capabilities the skill covers (scheduling, grants, cleanup); not 3 because it names more than 1-2 actions with specific parameters.

4 / 5

Completeness

Explicitly answers both: what ('managed ingestion into Unity Catalog from databases ... by CDC or query-based') and when ('Use when creating, running or debugging these pipelines'), with concrete trigger activities — closely matching the anchor-5 example structure.

5 / 5

Trigger Term Quality

Includes natural terms a Databricks user would say: 'Lakeflow Connect', 'ingestion', 'Unity Catalog', 'CDC', 'query-based', and the four database names. Not 5 because a few common synonyms users might use ('sync', 'replicate', 'ingest data from') are absent.

4 / 5

Distinctiveness Conflict Risk

Clear niche: product-named (Lakeflow Connect) database ingestion with enumerated sources; the SaaS/Kafka and file-based cases are explicitly routed to sibling skills in the body, so trigger overlap risk is minimal.

5 / 5

Total

18

/

20

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.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

15

/

16

Passed

Repository

databricks/databricks-agent-skills

Path

plugins/databricks/claude/skills/databricks-lakeflow-connect/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.