postgres-hybrid-text-search

Use this skill to implement hybrid search combining BM25 keyword search with semantic vector search using Reciprocal Rank Fusion (RRF). **Trigger when user asks to:** - Combine keyword and semantic search - Implement hybrid search or multi-modal retrieval - Use BM25/pg_textsearch with pgvector together - Implement RRF (Reciprocal Rank Fusion) for search - Build search that handles both exact terms and meaning **Keywords:** hybrid search, BM25, pg_textsearch, RRF, reciprocal rank fusion, keyword search, full-text search, reranking, cross-encoder Covers: pg_textsearch BM25 index setup, parallel query patterns, client-side RRF fusion (Python/TypeScript), weighting strategies, and optional ML reranking.

Quality

88%

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

77%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.

Highly actionable with clear sequencing and a verification/feedback loop, but held back by some marketing-style verbosity and a monolithic structure that keeps advanced material inline instead of progressively disclosed via reference files.

Suggestions

Trim marketing-style prose (e.g. the 'fully open-source and available hosted on Tiger Cloud' sentence) to only what Claude needs to act.

Move advanced material — pgvectorscale scaling and ML reranking — into reference files (e.g. references/SCALING.md, references/RERANKING.md) with one-level-deep signaled links from the main body.

Cross-check version-sensitive claims (e.g. 'pg_textsearch currently supports PostgreSQL 17 and 18') against the compatibility note, and isolate them so they do not bloat the core guide.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with executable code, but includes unnecessary prose such as 'pg_textsearch is a new BM25 text search extension ... fully open-source and available hosted on Tiger Cloud' (marketing-style context Claude does not need) that could be trimmed.	2 / 3
Actionability	Provides fully executable, copy-paste-ready SQL, Python, and TypeScript (RRF fusion, weighting, cross-encoder and Cohere reranking) with concrete parameters and operators.	3 / 3
Workflow Clarity	Golden Path sequences the steps (enable extensions, table + indexes, parallel queries, fuse, optional rerank) and the Monitoring & Debugging section supplies explicit EXPLAIN verification and an error-recovery feedback loop ('If EXPLAIN still shows sequential scans ... verify indexes exist').	3 / 3
Progressive Disclosure	Single ~270-line monolithic file with no bundle files in references/scripts/assets; advanced sections (pgvectorscale scaling, ML reranking) are inline rather than split into one-level-deep reference files, so structure is adequate but not optimally disclosed.	2 / 3
	Total	10 / 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.

A strong description: concrete actions, explicit trigger guidance, natural keyword coverage, and a distinct niche. Voice is the standard imperative skill form, with no first/second-person penalty.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'combining BM25 keyword search with semantic vector search using Reciprocal Rank Fusion (RRF)', 'pg_textsearch BM25 index setup, parallel query patterns, client-side RRF fusion (Python/TypeScript), weighting strategies, and optional ML reranking'.	3 / 3
Completeness	Clearly answers both what (implement hybrid search with RRF; covers indexing, query patterns, fusion, weighting, reranking) and when via an explicit 'Trigger when user asks to:' clause plus a Keywords line.	3 / 3
Trigger Term Quality	Good coverage of natural terms users would say — 'hybrid search', 'BM25', 'RRF', 'keyword search', 'full-text search', plus explicit trigger bullets like 'Combine keyword and semantic search' and 'Implement hybrid search'.	3 / 3
Distinctiveness Conflict Risk	Clear niche (BM25+vector hybrid search via RRF on pg_textsearch/pgvector) with distinct triggers; unlikely to fire for the wrong skill, though it lightly overlaps the referenced pgvector-semantic-search 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/postgres-hybrid-text-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.