Hyperspell icon

Hyperspell

Add documents to Hyperspell and run semantic search or grounded generation from any workflow

Actions10

Search → Answer

AI-generated

Summary

Run a natural-language query against Hyperspell's indexed data and generate a grounded answer with cited source documents.

Inputs

  • Query (required) — The natural-language query to run against Hyperspell.
  • Sources — Select specific sources to search (e.g., Vault, Gmail, Slack). Leave empty to search all sources, including Vault and every connected integration.
  • Simplify — Choose whether to return a simplified version of the response instead of raw data (default is true).
  • Options — Optional settings: Max Results limits the number of documents returned (default 10, max 200). Effort controls the amount of query pre-processing before retrieval, from Minimal (fastest, no LLM rewrite) to High (maximum LLM processing).
  • Act as User — Specify the user ID to run the request as, overriding the credential's user. Leave empty to use the credential's value or default app user.

Output shape

a list of source documents relevant to the query with generated answer text, each document containing grounding information including resource IDs, sources, and metadata.

The response may be simplified according to the Simplify parameter. The Max Results option limits the number of documents returned. Errors in API response are surfaced as node errors. If Act as User is not set, the request runs as the credential user or defaults to the app user with most documents.

Links

Discussion