hybrid-search
Use when building search systems that need both semantic similarity and keyword matching - covers combining vector and BM25 search with Reciprocal Rank Fusion, alpha tuning for search weight control, and optimizing retrieval quality
npx skills add majiayu000/claude-skill-registry --skill hybrid-search --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# LLMemory Hybrid Search ## Installation ```bash uv add llmemory # or pip install llmemory ``` ## Overview Hybrid search combines **vector similarity search** (semantic understanding) with **full-text search** (keyword matching) to deliver superior retrieval quality. Results are merged using **Reciprocal Rank Fusion (RRF)** to create a unified ranking. **When to use hybrid search:** - Need both semantic similarity AND exact keyword matches - Queries contain specific terms, names, or technical jargon - Want best-of-both-worlds retrieval quality (recommended default) **When to use vector-only search:** - Purely semantic/conceptual queries - Cross-lingual search - Queries with synonyms or paraphrasing **When to use text-only search:** - Exact keyword/phrase matching required - Search in structured data or code - When embeddings are not available ## Quick Start ```python from llmemory import LLMemory, SearchType async with LLMemory(connection_string="postgresql://localhost/mydb") as memory: # Hybrid search (default, recommended) results = await memory.search( owner_id="workspace-1", query_text="machine learning algorithms", search_type=SearchType.HYBRID, limit=10, alpha=0.5 # Equal wei
- Installation
- Overview
- Quick Start
- Complete API Documentation
- SearchType Enum
- search() - Hybrid Mode
- Understanding Alpha Parameter
- Alpha Values Guide
- Choosing Alpha for Different Query Types
- Reciprocal Rank Fusion (RRF)
- How RRF Works
- RRF Benefits
- Example: RRF in Action
- Configuring Hybrid Search with SearchConfig
uv add llmemory or pip install llmemory Use fast profile for low-latency applications export LLMEMORY_HNSW_PROFILE=fast Use accurate profile for high-precision requirements export LLMEMORY_HNSW_PROFILE=accurate Use balanced profile (default, can be omitted) export LLMEMORY_HNSW_PROFILE=balanced Lower k favors top-ranked results
What does the hybrid-search skill do?
Use when building search systems that need both semantic similarity and keyword matching - covers combining vector and BM25 search with Reciprocal Rank Fusion, alpha tuning for search weight control, and optimizing retrieval quality
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill hybrid-search --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.