Gemini File Search: Query (Tool) icon

Gemini File Search: Query (Tool)

Query a Gemini File Search store as a Tool for AI Agents

Overview

This node allows querying a Gemini File Search store using a specified AI model to answer questions based on the store's content. It is useful for integrating AI-powered search and question-answering capabilities into workflows, enabling users to retrieve information from file stores efficiently. For example, it can be used to query documentation, knowledge bases, or any indexed file store to get precise answers to user questions.

Use Case Examples

  1. Query a file store containing company policies to answer employee questions.
  2. Use the node to fetch specific information from a large document repository by asking natural language questions.

Properties

Name Meaning
Store Name The full resource name of the Gemini File Search store to query.
Question The question to be answered using the File Search store.
Model The AI model used to answer the question.
Include Citations Whether to include citations (grounding metadata) in the response.

Output

JSON

  • text - The answer text generated by the AI model based on the query.
  • groundingMetadata - Optional citations or metadata grounding the answer, included if requested and available.

Dependencies

  • Requires an API key credential for the Gemini API to authenticate requests.

Troubleshooting

  • Ensure the Gemini API credential is correctly configured and has necessary permissions.
  • Verify the store name is accurate and accessible.
  • Check network connectivity to the Gemini API endpoint.
  • If no candidates are returned, the store might not contain relevant information or the question might be too ambiguous.

Links

Discussion