vector-search

Vector Search

HNSW-indexed vector search with RuVector embeddings (384-dim ONNX).

When to use

When you need fast semantic search across large knowledge bases. HNSW provides 150x-12,500x faster search compared to brute-force scanning.

Steps

1. Check status — call mcp__claude-flow__embeddings_status to verify the embedding engine
2. Initialize — call mcp__claude-flow__embeddings_init if not already active
3. Generate embeddings — call mcp__claude-flow__embeddings_generate for text input
4. Search — call mcp__claude-flow__embeddings_search with a query for semantic matches
5. Compare — call mcp__claude-flow__embeddings_compare to measure similarity between texts
6. Unified search — call mcp__claude-flow__memory_search_unified to search across all namespaces

HNSW Index

For building custom HNSW indexes:

• mcp__claude-flow__ruvllm_hnsw_create — create a new index
• mcp__claude-flow__ruvllm_hnsw_add — add vectors to the index
• mcp__claude-flow__ruvllm_hnsw_route — route queries through the index

Hyperbolic embeddings

For hierarchical data (code trees, org charts), use mcp__claude-flow__embeddings_hyperbolic which maps to Poincare ball space.

CLI alternative

npx @claude-flow/cli@latest embeddings search --query "authentication patterns"
npx @claude-flow/cli@latest embeddings init
npx @claude-flow/cli@latest memory search --query "your query"

Performance

Method	Speed
Brute-force scan	Baseline
HNSW (n=500)	150x faster
HNSW (n=10,000)	12,500x faster