createVectorQueryTool()
The createVectorQueryTool() function creates a tool for semantic search over vector stores. It supports filtering, reranking, and integrates with various vector store backends.
Basic Usage
import { createVectorQueryTool } from "@mastra/rag";
const queryTool = createVectorQueryTool({
vectorStoreName: "pinecone",
indexName: "docs",
options: {
provider: "OPEN_AI",
model: "text-embedding-ada-002",
maxRetries: 3
}
});Parameters
vectorStoreName:
string
Name of the vector store to query (must be configured in Mastra)
indexName:
string
Name of the index within the vector store
options:
EmbeddingOptions
Configuration for embedding generation
topK?:
number
Maximum number of results to retrieve
vectorFilterType?:
'pg' | 'astra' | 'qdrant' | 'upstash' | 'pinecone' | 'chroma' | ''
Type of vector store for filter formatting
rerankOptions?:
RerankerOptions
Options for reranking results
Returns
The tool returns an object with:
relevantContext:
string
Combined text from the most relevant document chunks
Example with Filters
// Pinecone/PG/Astra
const queryTool = createVectorQueryTool({
vectorStoreName: "pinecone",
indexName: "docs",
options: {
provider: "OPEN_AI",
model: "text-embedding-ada-002",
maxRetries: 3
},
vectorFilterType: "pinecone",
topK: 5
});Filter Formats:
- Pinecone/PG/Astra:
{ category: { eq: "technical" } }
Example with Reranking
const queryTool = createVectorQueryTool({
vectorStoreName: "milvus",
indexName: "documentation",
options: {
provider: "OPEN_AI",
model: "text-embedding-ada-002",
maxRetries: 3
},
topK: 5,
rerankOptions: {
model: "cross-encoder",
threshold: 0.7
}
});Tool Details
The tool is created with:
- ID:
VectorQuery {vectorStoreName} {indexName} Tool - Description:
Fetches and combines the top {topK} relevant chunks from the {vectorStoreName} vector store using the {indexName} index - Input Schema: Requires queryText and filter objects
- Output Schema: Returns relevantContext string