CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

databricks-vector-search

Databricks AI Search (formerly Vector Search) endpoints and indexes for RAG and semantic search. Use when asked to: create Standard or Storage-Optimized endpoints and Delta Sync or Direct Access indexes with AISearchClient; query with ANN, hybrid, or full-text search and filters; sync, monitor, or migrate indexes; set a target QPS, triage 429s, or load-test an endpoint; wire an index into an agent as a retriever tool. NOT for: no-code document Q&A (databricks-genie-agents); serving-endpoint CRUD (databricks-model-serving); custom ResponsesAgent authoring and deployment (databricks-ml-training).

Quality

88%

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

81%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, exceptionally actionable body with executable examples and well-validated multi-step workflows, backed by a real, well-signaled reference bundle. Its main weakness is conciseness: scattered version archaeology and duplicated content that a dedicated compatibility/operations reference could absorb.

Suggestions

Move the legacy databricks-vectorsearch compatibility notes and the 0.68-0.82 version-history details into a dedicated 'Version compatibility / old patterns' section or into references/troubleshooting-and-operations.md, leaving only the minimum supported version inline.

Drop the 'When to Use' section — it duplicates the frontmatter description's trigger list and adds no new information.

Trim the Production and High QPS section to a short checklist inline and move the route/pitfall deep-dive prose to references/troubleshooting-and-operations.md.

DimensionReasoningScore

Conciseness

The body is mostly dense, high-value gotchas, but version-number history is scattered inline rather than in a deprecated/old-patterns section (an ~11-line legacy databricks-vectorsearch paragraph plus 0.79/0.80/0.81/0.82 mentions throughout), the 'When to Use' section duplicates the frontmatter triggers, and the Notes section repeats the filter-syntax rule verbatim. More than minor trimming is needed, so it sits at anchor 3 rather than 4.

3 / 5

Actionability

Fully executable, copy-paste-ready code and commands throughout: create_endpoint/create_delta_sync_index variants with real parameters, both filter syntaxes, the scan pagination loop, the target_qps polling loop with int() handling of string int64 fields, and complete CLI invocations. Matches the top anchor.

5 / 5

Workflow Clarity

Multi-step processes carry explicit validation checkpoints and error-recovery feedback: check endpoint state before retrying creation, wait for a running update instead of re-triggering sync, the three separate waits (ONLINE / SCALING_CHANGE_APPLIED / status.ready) with 'Load-test only after all three', and 429 triage, plus an ask-first safety gate for costly/destructive operations. Matches the top anchor.

5 / 5

Progressive Disclosure

The four reference files all exist, are one level deep, and are well signaled via the annotated Reference Files table plus an inline link in the hybrid-search section; however, the long Production and High QPS deep-dive prose and the legacy-version compatibility material are inlined in SKILL.md where they could live in troubleshooting-and-operations.md. Good structure with minor placement gaps — anchor 4, not 5.

4 / 5

Total

17

/

20

Passed

Description

92%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 and comprehensive capabilities, explicit trigger guidance, and explicit negative boundaries against sibling skills. The only weakness is a couple of missing common synonyms such as 'similarity search' and 'embeddings' among the trigger terms.

DimensionReasoningScore

Specificity

The description enumerates multiple concrete, comprehensive actions with the named client — 'create Standard or Storage-Optimized endpoints and Delta Sync or Direct Access indexes with AISearchClient', 'query with ANN, hybrid, or full-text search and filters', 'set a target QPS, triage 429s, or load-test an endpoint', 'wire an index into an agent as a retriever tool' — matching the top anchor with no meaningful coverage gaps.

5 / 5

Completeness

It clearly answers both 'what' (first sentence describing endpoints and indexes for RAG and semantic search) and 'when' (explicit 'Use when asked to:' clause with concrete trigger phrases), matching the top anchor.

5 / 5

Trigger Term Quality

Good natural keyword coverage ('RAG', 'semantic search', 'formerly Vector Search', 'filters', '429s', 'load-test', 'retriever tool'), but a few common user phrasings like 'similarity search' and 'embeddings' are missing, keeping it just below the comprehensive anchor.

4 / 5

Distinctiveness Conflict Risk

A named product niche plus an explicit 'NOT for:' boundary listing three sibling skills makes conflict with other skills minimal, matching the clear-niche top anchor.

5 / 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: 4 suspicious

Warning

Total

14

/

16

Passed

Repository

databricks/databricks-agent-skills

Path

plugins/databricks/claude/skills/databricks-vector-search/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.