hybrid-search-implementation

Combine vector and keyword search for improved retrieval. Use when implementing RAG systems, building search engines, or when neither approach alone provides sufficient recall.

1.17x

Quality

66%

Does it follow best practices?

Impact

95%

1.17x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./plugins/llm-application-dev/skills/hybrid-search-implementation/SKILL.md

Evaluation results

98%

11%

Search Result Fusion Library

RRF and linear fusion algorithms

Criteria

Without context

With context

RRF constant k=60

100%

RRF rank offset

100%

RRF score accumulation

100%

Linear normalization

100%

Linear alpha formula

60%

90%

Linear alpha default

50%

83%

Union of doc sets

100%

Empty result handling

100%

Descending sort

100%

Result format

100%

Zero division guard

100%

Weighted RRF support

100%

10%

Knowledge Base Search for a Support Platform

PostgreSQL hybrid search schema

Criteria

Without context

With context

pgvector extension

100%

HNSW vector index

100%

Cosine ops index

100%

GIN full-text index

100%

Generated tsvector column

100%

tsvector language

100%

Cosine distance operator

100%

websearch_to_tsquery

100%

RRF fusion in SQL

60%

100%

Both scores returned

100%

FULL OUTER JOIN

100%

asyncpg usage

100%

88%

20%

Retrieval Component for a Medical Literature Assistant

Hybrid RAG pipeline with reranking

Criteria

Without context

With context

SearchResult dataclass

75%

100%

Source field values

100%

Parallel async search

100%

Candidate pool multiplier

100%

Cross-encoder model

Reranker integration

100%

RRF as default fusion

100%

RRF k=60

100%

Empty candidates guard

100%

Final result limiting

100%

Metadata preserved

100%

Configurable fusion method

100%

Repository: wshobson/agents
Commit: 91fe43e

Evaluated: 28 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Search Result Fusion Library Knowledge Base Search for a Support Platform Retrieval Component for a Medical Literature Assistant

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.