Entity Enricher icon

Entity Enricher

Enrich entities with multi-model LLM fusion, multilingual output, and expertise-driven strategies

Actions18

Schema → Generate Schema

AI-generated

Summary

Generate and auto-save a JSON schema representing the union of fields across all input items, which must be samples of the same entity type.

Inputs

  • Authentication — Choose the authentication method to use: API Key or OAuth2.
  • Connected To Name or ID — Select the connected organization and credential or specify an ID via expression.
  • Model — Schema generation model key or 'auto' to use the organization's default model.
  • Generate Semantic IDs — Boolean flag to add semantic_id properties to objects for embedding-based entity resolution.
  • Commonality Threshold — A number between 0.05 and 1 indicating the minimum share of fields every input item must have to avoid early failure.
  • Language — Optional language code used to describe the schema itself; sample property names are always taken from the input samples.
  • Timeout (Seconds) — Maximum time in seconds to wait for schema generation to complete.

Output shape

a single record

Returns a JSON object detailing success, the generated schema ID and schema content, suggestions, counts of properties and expertise, model and cost metadata, and processing times. The operation runs once for the entire batch of input items, which must not be empty and represent samples of the same entity type.

Examples

Example 1: Generate a JSON schema for a given set of input sample entities

Input items contain JSON objects each representing one sample of the same entity type. Configure authentication and optionally select the model and parameters to control schema generation. Upon execution, the node generates a JSON schema that covers the union of all fields seen across samples, marks missing or null fields as nullable, and uses distinct observed values as examples. The schema is automatically saved in the connected Entity Enricher organization.

Links

Discussion