Function getMatchingNodesWithVectorSearch

  • getMatchingNodesWithVectorSearch(db, query, limit?): Promise<{
        context: string;
        eids: string[];
        embedding: number[];
        nodes: NodeResult[];
        searchResults: node.introspection.query.SimilarityResult[];
    }>

  • Complete vector search with context generation This combines all steps: embedding generation, search, node retrieval, and context formatting

    Parameters

    • db: Surreal

      SurrealDB instance

    • query: string

      Natural language query text

    • limit: number = 5

      Number of results to return (default: 5)

    Returns Promise<{
        context: string;
        eids: string[];
        embedding: number[];
        nodes: NodeResult[];
        searchResults: node.introspection.query.SimilarityResult[];
    }>

    Object with all search data including:

    • embedding: The generated query embedding vector
    • searchResults: Raw search results with distance scores and nodes
    • eids: Array of embedding cache record IDs
    • nodes: Array of matched node data
    • context: Formatted string ready for LLM consumption

    Example

    const result = await getMatchingNodesWithVectorSearch(db, "authentication functions", 5);
    console.log(result.context); // LLM-ready context string
    console.log(result.nodes);   // Raw node data

Settings

Member Visibility