databricks-vector-search

Patterns for Databricks Vector Search: create endpoints and indexes, query with filters, manage embeddings. Use when building RAG applications, semantic search, or similarity matching. Covers both storage-optimized and standard endpoints.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

65%

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

A thorough, actionable reference for Databricks Vector Search with strong executable examples, but it is long and repetitive across SDK/MCP/CLI surfaces and lacks validation checkpoints for destructive and async operations. Its referenced detail files are not present in the bundle.

Suggestions

Add explicit validation/feedback loops: poll endpoint status after create_endpoint until ACTIVE, and verify index sync success after delete/upsert operations before proceeding.

Move the large MCP-tools and CLI sections into reference files (or consolidate SDK vs MCP examples) to reduce repetition and shrink the inline body.

Either include the referenced bundle files (search-modes.md, end-to-end-rag.md, index-types.md, troubleshooting-and-operations.md) in the skill package or remove the dangling links from the Reference Files table.

Dimension	Reasoning	Score
Conciseness	The body is mostly dense reference material with executable code, but the same operations (endpoint creation, querying) recur across the SDK, MCP, and CLI sections, and the MCP-tools block is large enough to feel like inline duplication.	2 / 3
Actionability	Code blocks are executable and copy-paste ready across the Databricks SDK, MCP tools, and CLI, with concrete parameter values and result-handling examples.	3 / 3
Workflow Clarity	Operations are listed clearly, but destructive actions (delete index, delete data) and asynchronous endpoint creation lack explicit validation or poll-until-ready feedback loops, which caps this dimension at 2.	2 / 3
Progressive Disclosure	A Reference Files table signals one-level-deep links, but none of the referenced files (search-modes.md, end-to-end-rag.md, etc.) exist in any bundle directory, and most detail (MCP tools, filters, CLI) remains inline rather than being split out.	2 / 3
	Total	9 / 12 Passed

Description

100%

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 concise, well-targeted description that names concrete capabilities, includes a clear "Use when" trigger, and occupies a distinct niche. It uses third-person voice without over-claims or fluff.

Dimension	Reasoning	Score
Specificity	"create endpoints and indexes, query with filters, manage embeddings" lists multiple concrete, distinct actions rather than vague language.	3 / 3
Completeness	It states both what it does (create/query/manage) and when to use it via an explicit "Use when building RAG applications, semantic search, or similarity matching" clause.	3 / 3
Trigger Term Quality	"Databricks Vector Search", "RAG applications", "semantic search", and "similarity matching" are natural terms users would actually say when they need this skill.	3 / 3
Distinctiveness Conflict Risk	Scoped to Databricks Vector Search with distinctive triggers (RAG, semantic search, storage-optimized vs standard), making overlap with other skills unlikely.	3 / 3
	Total	12 / 12 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
relative_links	Relative link issues: 6 missing, 5 suspicious	Warning

	Total	15 / 16 Passed

Repository: databricks-solutions/ai-dev-kit
Commit: 1c43c21

Reviewed: 3 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.