Skip to content
Nappai Documentation

Upstash

This component lets you search and store information within Nappai using Upstash, a fast and efficient vector database. Think of it as a highly advanced search tool that understands the meaning of your data, not just keywords. You can add new information and then search for similar items based on their content.

Relationship with Upstash

This component directly interacts with Upstash’s vector database. It uses Upstash’s API to store and retrieve information, making it easy to search for similar pieces of data within your Nappai workflows.

Inputs

  • Text Key: This specifies which part of your data should be used for searching. By default, it uses the “text” field. You can change this to a different field if needed (advanced users).
  • Search Query: This is what you type to search for similar information. Use clear and concise language.
  • Ingest Data: This is the information you want to add to the Upstash database. It should be a list of items.
  • Embedding: This lets you specify how your data is represented for searching. Usually, you don’t need to change this; Upstash will handle it automatically.
  • Number of Results: This controls how many search results are returned. The default is 4. You can change this to get more or fewer results (advanced users).
  • Credential: This is your Upstash API key, which allows Nappai to access your Upstash account. Keep this secure!

Outputs

The component returns a list of items that are similar to your search query. These items are ranked by how closely they match your search. You can use this output in other parts of your Nappai workflow to process or display the results.

Usage Example

Imagine you have a database of customer reviews. You can use this component to:

  1. Ingest Data: Add all your customer reviews to the Upstash database.
  2. Search Query: Type a search query like “excellent customer service”.
  3. Get Results: The component will return a list of reviews that mention excellent customer service, ranked by relevance.
  4. Use Results: You can then use these results to analyze customer sentiment or display them on a dashboard.

Templates

[List of templates where the component is used. This section will be populated based on the actual templates.]

  • VectorStoreInfo: Provides information about the vector store being used.
  • Self Query Retriever: Uses this component to search a vector store and generate queries.
  • [Other related components]: [Brief descriptions of other relevant components from the provided list. Select only the most relevant ones and provide concise descriptions.] For example: Google Sheet Writer could be used to export the search results to a Google Sheet.

Tips and Best Practices

  • Use clear and concise search queries for better results.
  • Ensure your ingested data is well-structured and relevant to your search needs.
  • Regularly review and update your Upstash credentials to maintain security.

Security Considerations

  • Protect your Upstash API key: Never share your Upstash credential with anyone. Treat it like a password.
  • Regularly audit access: Review who has access to your Upstash account and the Nappai workflow using this component.