Stats
Actions
Tags
From llm-application-dev
Combines vector similarity and keyword search for improved recall in RAG systems and search engines. Covers fusion methods like RRF, linear scoring, and cross-encoder reranking.
How this skill is triggered — by the user, by Claude, or both
Slash command
/llm-application-dev:hybrid-search-implementationThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Patterns for combining vector similarity and keyword-based search.
Patterns for combining vector similarity and keyword-based search.
Query → ┬─► Vector Search ──► Candidates ─┐
│ │
└─► Keyword Search ─► Candidates ─┴─► Fusion ─► Results
| Method | Description | Best For |
|---|---|---|
| RRF | Reciprocal Rank Fusion | General purpose |
| Linear | Weighted sum of scores | Tunable balance |
| Cross-encoder | Rerank with neural model | Highest quality |
| Cascade | Filter then rerank | Efficiency |
Full template library and detailed worked examples live in references/details.md. Read that file when you need the concrete templates.
npx claudepluginhub yo-steven/agents-exploration-20260523 --plugin llm-application-devImplements hybrid search combining vector and keyword retrieval for RAG systems, search engines, and domain-specific queries requiring high recall.
Combines vector similarity and keyword search for improved retrieval in RAG systems and search engines.
Hybrid search — BM25 keyword + dense vector fused with Reciprocal Rank Fusion. Use when queries have specific identifiers, code symbols, or proper nouns that pure semantic search might miss.