Overview

This node performs a search operation on the Moorcheh Vector Store API. It allows users to search for relevant documents or vectors within specified namespaces using either a text query or a vector query (comma-separated vector values). The search can be customized with parameters such as the number of top results to return (Top K), a threshold for filtering results, and a kiosk mode toggle that adjusts the search behavior. This node is useful for applications involving semantic search, vector similarity search, or retrieving relevant information from a vector database.

Use Case Examples

1. Searching for documents related to a specific text query across multiple namespaces.
2. Performing a vector similarity search by providing a vector query to find the closest matching vectors in the database.
3. Filtering search results by setting a threshold to exclude less relevant matches.
4. Using kiosk mode to enable additional filtering based on the threshold parameter.

Properties

Output

JSON

• json - The JSON response from the Moorcheh API containing the search results.

Dependencies

• Moorcheh API key credential for authentication

Troubleshooting

• Ensure the query format matches the expected input: a text string or comma-separated vector values with correct dimension.
• Verify that the namespaces specified exist and are accessible in the Moorcheh vector store.
• If kiosk mode is enabled, check that the threshold value is set appropriately to avoid filtering out all results.
• Common error messages may include invalid JSON in metadata, unsupported operations, or authentication failures. Verify the API key and request parameters.