pgvector-semantic-search

Use this skill for setting up vector similarity search with pgvector for AI/ML embeddings, RAG applications, or semantic search. **Trigger when user asks to:** - Store or search vector embeddings in PostgreSQL - Set up semantic search, similarity search, or nearest neighbor search - Create HNSW or IVFFlat indexes for vectors - Implement RAG (Retrieval Augmented Generation) with PostgreSQL - Optimize pgvector performance, recall, or memory usage - Use binary quantization for large vector datasets **Keywords:** pgvector, embeddings, semantic search, vector similarity, HNSW, IVFFlat, halfvec, cosine distance, nearest neighbor, RAG, LLM, AI search Covers: halfvec storage, HNSW index configuration (m, ef_construction, ef_search), quantization strategies, filtered search, bulk loading, and performance tuning.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

65%Weight 40%Scale 1-3

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

A highly actionable, well-structured pgvector reference with executable SQL throughout, but it is verbose in places and monolithic with no progressive disclosure into bundle files. Validation checkpoints are present but not framed as explicit feedback loops for batch operations.

Suggestions

Trim the conceptual opening paragraph and consolidate the repeated ef_search/memory tuning guidance into one section to improve conciseness.

Add an explicit validate→fix→retry checkpoint sequence around the bulk-load and index-build flow (e.g., verify row count, check index build progress, EXPLAIN ANALYZE before declaring done).

Split the detailed quantization and filtering-strategy sections into reference files under references/ and link to them one level deep to apply progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Mostly efficient dense reference content, but the opening paragraph explains concepts Claude already knows ('Semantic search finds content by meaning...') and ef_search/memory tuning recurs across Golden Path, HNSW Parameters, and Performance sections — could be tightened.	2 / 3
Actionability	Provides fully executable, copy-paste-ready SQL throughout (Standard Pattern, HNSW/IVFFlat creation, binary-quantization re-ranking query, bulk loading, monitoring), with concrete parameters matching the top anchor.	3 / 3
Workflow Clarity	Sequenced guidance and validation probes exist (EXPLAIN ANALYZE, recall-comparison BEGIN/COMMIT), but for batch/database operations like bulk loading and index creation there is no explicit validate→fix→retry checkpoint loop, which caps clarity per the rubric.	2 / 3
Progressive Disclosure	Well-organized into clear sections, but it is a monolithic ~345-line file with no local bundle files; detailed reference material (quantization tables, filtering strategies) that could be split into separate files is kept inline.	2 / 3
	Total	9 / 12 Passed

Description

100%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, well-triggered, and complete, naming concrete capabilities and explicit 'when to use' guidance for a distinct pgvector niche. No significant weaknesses; the conventional imperative 'Use this skill' phrasing matches accepted examples.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'setting up vector similarity search', 'Store or search vector embeddings', 'Create HNSW or IVFFlat indexes', 'Use binary quantization' — matching the top anchor for specific concrete actions.	3 / 3
Completeness	Clearly answers both what (pgvector vector similarity search; 'Covers: halfvec storage, HNSW index configuration...') and when via an explicit 'Trigger when user asks to' clause, satisfying the top anchor.	3 / 3
Trigger Term Quality	An explicit 'Trigger when user asks to' list plus a 'Keywords' line covers natural terms users would say (semantic search, nearest neighbor, RAG, embeddings, cosine distance) with good variation.	3 / 3
Distinctiveness Conflict Risk	A clear pgvector-specific niche with distinct triggers (HNSW, IVFFlat, halfvec) makes it unlikely to fire for the wrong skill.	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
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: timescale/pg-aiguide
Path: skills/pgvector-semantic-search/SKILL.md
Commit: b4f11a4

Reviewed: 11 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.