Tessl • The package manager for agent skills

SkillPublic

databricks-vector-search

Databricks Vector Search endpoints and indexes for RAG and semantic search; covers index types, search modes, end-to-end RAG patterns

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./plugins/databricks/claude/skills/databricks-vector-search/SKILL.md

Quality

Content

68%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 well-structured, highly actionable reference: executable SDK/CLI examples, useful comparison tables, and a clean one-level-deep reference bundle. The main weaknesses are the absence of validation around destructive index/data deletion and minor verbosity plus inline detail that belongs in the reference files.

Suggestions

Add validation before destructive operations — e.g. check get-index status and confirm the index name before delete-index / delete_data_vector_index, or require an explicit user confirmation step.

Move the filter-syntax examples and embedding-model table into the existing reference files and keep only a one-line pointer plus the Standard-vs-Storage-Optimized distinction inline.

Replace placeholder ellipsis vectors ([0.1, 0.2, ...]) with a generated 768-dim example or a note on constructing them, so examples are truly copy-paste runnable.

Dimension	Reasoning	Score
Conciseness	The body is dense and efficient — tables for comparisons, tight code blocks, no tutoring on concepts Claude already knows — but has minor padding: comments duplicating adjacent prose ('# Combines vector similarity with keyword matching') and overlap between the Common Issues table and the Notes section.	4 / 5
Actionability	Code is concrete and near copy-paste ready with real argument names and values across create/query/filter/sync flows, but placeholder vectors like [0.1, 0.2, ...] and the unexplained doc[-1] score indexing leave minor gaps versus fully executable.	4 / 5
Workflow Clarity	Quick Start sequences create endpoint → create index → query with an async-status checkpoint and Common Issues aids recovery, but destructive operations (delete-index, delete_data_vector_index) appear with no validation or confirmation steps, which caps workflow clarity at 3 per the destructive-operations rule.	3 / 5
Progressive Disclosure	SKILL.md works as an overview with four real, one-level-deep reference files clearly signaled in a table with descriptions (all four linked paths exist in references/), but a fair amount of detail (filter syntax examples, embedding-model tables) is inlined that could live in the reference files.	4 / 5
	Total	15 / 20 Passed

Description

70%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 fairly strong description: specific, distinctive, and keyword-rich, but it omits any explicit 'when to use this skill' trigger clause, which both caps completeness and weakens its usefulness for skill triggering. Adding a 'Use when...' sentence would lift it to the top band.

Suggestions

Add an explicit trigger clause, e.g. 'Use when building RAG or semantic search on Databricks, creating vector indexes from Delta tables, or choosing between Standard and Storage-Optimized endpoints.'

Include common user synonyms such as 'similarity search', 'embeddings', and 'retrieval-augmented generation' alongside 'RAG'.

Frame capabilities as concrete actions ('Create, manage, and query vector search endpoints and indexes') rather than the passive 'covers ... patterns'.

Dimension	Reasoning	Score
Specificity	The description names the domain and enumerates several specific topics — "endpoints and indexes", "index types, search modes, end-to-end RAG patterns" — but states scope rather than concrete actions (create/manage/query), so it falls between the 3 and 5 anchors.	4 / 5
Completeness	The 'what' is clear (Vector Search endpoints and indexes for RAG/semantic search, covering index types and search modes), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Good natural keywords users would actually say ("RAG", "semantic search", "vector search", "Databricks"), but common variations like "similarity search", "embeddings", and "retrieval" are missing, matching the 'good coverage, a few natural terms missing' anchor.	4 / 5
Distinctiveness Conflict Risk	"Databricks Vector Search" is a clear niche with distinct triggers (RAG, semantic search on Databricks) and minimal conflict risk; it would not be confused with sibling Databricks skills such as model-serving or unity-catalog.	5 / 5
	Total	16 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 3 suspicious	Warning

	Total	14 / 16 Passed

Repository: databricks/databricks-agent-skills
Path: plugins/databricks/claude/skills/databricks-vector-search/SKILL.md
Revision: c570f88

Reviewed: 7 days ago

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.

databricks-vector-search

Table of Contents

Is this your skill?