n8n Connector - Entity Enricher Documentation

n8n Connector

The Entity Enricher community node for n8n lets you build automated enrichment pipelines with a visual drag-and-drop editor. Connect to 400+ apps without writing code.

Installation

Install the community node in your n8n instance:

npm install n8n-nodes-entity-enricher

Or install via the n8n UI: go to Settings → Community Nodes and search for n8n-nodes-entity-enricher.

Prerequisites

API Key— Create an organization access key in Entity Enricher. See the API Keys documentation for details. The key format is ent_XXXXXXXXXXXX.

Base URL— The URL of your Entity Enricher instance (e.g., https://entityenricher.ai or your self-hosted URL).

Credential Setup— In n8n, create an Entity Enricher API credential with your API key and base URL. All nodes will use this credential. An organization access key acts independently of any user account — the recommendation for durable service-to-service workflows.

OAuth2 (alternative)— Prefer connecting with your account? Create an Entity Enricher OAuth2 API credential instead: copy the OAuth redirect URL n8n shows, register an OAuth client with it in Entity Enricher under API Keys → Connected Apps → OAuth Clients (owner role), paste the client ID into n8n, and click Connect my account. The connection acts with your own role and is revocable anytime under Connected Apps. Select the credential on the node with its Authentication parameter.

Available Operations

The connector exposes 18 operations across 7 resource categories:

Resource	Operation	Description
Enrichment	Enrich Entity	Enrich a single entity — just pick a schema; your organization's best model and strategy are chosen automatically. Streams results via SSE with real-time progress.
Enrichment	Enrich Entity Advanced	Enrich a single entity with full control over models, fusion, strategy, classification, and structured output.
Enrichment	Batch Enrich	Enrich all input items as a single batch with the automatic model and strategy. Parallel processing with per-entity progress tracking.
Enrichment	Batch Enrich Advanced	Enrich all input items as a single batch with the full parameter set.
Schema	List Schemas	List available saved schemas for selection in enrichment workflows.
Schema	Get Schema Details	Retrieve full schema content with extracted key properties for a specific schema.
Schema	Generate Sample	Generate 1..N realistic sample JSON objects of one entity type in a single job — the entry point of the schema-authoring loop. Language defaults to auto, inferred from the words of your own request.
Schema	Generate Schema	Generate and auto-save a JSON schema — every input item is one sample of the same entity type: union of fields, nullable where a field is missing in some sample, real observed values as examples. A Language option sets the language the schema describes itself in.
Record	List Records	Query enrichment records with type and success filters. Paginated results.
Record	Get Record	Retrieve a specific enrichment result by ID with full structured output.
Record	Sync Records to Database	Send a stored (or transformed) enrichment output to its schema's database sync — re-validated against the published contract, then passed through the admission gate.
Fusion	Merge Results	Fuse results from multiple enrichment records with optional LLM arbitration.
Attachment	Add Attachment	Upload one or more binary files from the input item (single multipart request) and return one item per attachment ID.
Attachment	Delete Attachment	Delete an attachment by ID — a handy post-enrichment cleanup step.
Database Sync	List Database Syncs	List the database syncs registered on a schema, with pending delta counts.
Database Sync	Fetch Deltas	Fetch the next FIFO window of database deltas (SQL + JSON), optionally leasing them for acknowledgement.
Database Sync	Acknowledge Deltas	Acknowledge applied deltas up to an ID — releases the lease and, per sync options, purges delivered copies.
Configuration	Get Options	Fetch available models, languages, strategies, and plan limits for dynamic configuration.

Trigger Node

The package also ships a separate Entity Enricher Trigger node — a webhook trigger that starts a workflow on schema events, registering its subscription automatically:

Enrichment Result— Fires on every completed enrichment of the selected schema (with or without a database sync).

Rejected for Database Save— Fires when an enrichment fails a database sync's admission gate — the payload tells you which required fields were missing.

Database Deltas Available— Fires when new SQL deltas are ready for a database sync. With "Fetch Deltas on Fire" enabled the node fetches and leases the pending deltas, emitting one item per delta for a downstream Acknowledge Deltas step. A database fed by several schemas notifies per schema — each drives its own workflow — so pick which linked schema this trigger listens to.

Together with the Database Sync operations, the trigger keeps an external database converged from n8n alone — see the Database Sync documentation.

Single Entity Enrichment

The simplest workflow: trigger on new data, enrich a single entity, and push the result to your destination. The node waits for the full enrichment to complete before passing the structured output downstream.

Example workflow — single entity enrichment:

Node configuration — Enrich Entity operation:

Configuration

The default Enrich Entity and Batch Enrich operations show only the essentials: schema, file upload, languages, and web search. Entity Enricher automatically runs with your organization's best model — your pinned default, or the top benchmark-scored model — and picks the enrichment strategy from your schema. The Enrich Entity Advanced and Batch Enrich Advanced operations expose the full set of options below.

Select a schema from the dynamic dropdown, choose one or more models, and provide entity data as key-value fields. The node auto-populates search key fields based on the selected schema.

Optional settings include strategy (single-pass or multi-expertise), languages for multilingual output, classification model for entity type verification, arbitration model for multi-model conflict resolution, and response schema / strict structured output toggles that enforce schema-conformant output on capable models.

To feed source documents into the enrichment, toggle Upload Input Binary Files: the node uploads the input item's binary files as attachments in a single request (all of them, or just the ones named in Binary Fields to Upload) and, with Delete Uploaded Attachments After Enrichment (default on), cleans them up once the enrichment finishes — no separate Add Attachment / Delete Attachment steps needed. Alternatively, supply an Attachment IDs field (comma-separated UUIDs from prior Add Attachment calls) to reuse documents uploaded once across many enrichments; both sources can be combined. Also available on Batch Enrich, where files are gathered from all input items and apply to every entity in the job. If an Edit Fields node sits between the file source and the enrich node, enable its Include Other Input Fields switch — otherwise it strips binary data.

Batch Enrichment

Process all input items as a single batch. Ideal for enriching lists from spreadsheets, databases, or API responses. All entities are processed in parallel with per-provider rate limiting.

Node configuration — Batch Enrich operation:

Entity Enricher node configuration for batch enrichment

How Batch Mode Works

Collect items — The node gathers all input items from the previous node (e.g., rows from a spreadsheet).
Start batch job — All entities are submitted via POST /api/batch/start and processed in parallel.
Stream progress — The node consumes the SSE stream until all entities complete or the timeout is reached.
Output results — One output item per entity with the enriched structured data, ready for downstream processing.

Key Features

Simple & Advanced Operations

The default Enrich Entity and Batch Enrich operations show just schema, file upload, languages, and web search — the best model and strategy are picked automatically. The Advanced operations give full control over models, fusion, strategy, and structured output.

Dynamic Dropdowns

Schema, model, language, and strategy selectors fetch options from your Entity Enricher instance at configuration time.

SSE Streaming

Enrichment operations consume the SSE stream in real-time, waiting for completion before returning results to the next node.

Auto-Continue

When pre-flight classification detects a mismatch, the connector automatically continues (n8n workflows are non-interactive).

Configurable Timeout

Default 300-second timeout per enrichment call, adjustable per node for large schemas or slow models.

Multi-Model Support

Select multiple models per enrichment. When 2+ models are used, results are automatically fused.

Clean Output by Default

By default, the output contains only the enriched data fields. Toggle "Include Enrichment Metadata" to add cost, tokens, fusion details, and record IDs.

Per-Model Output

Optionally output individual model results alongside the fused result for downstream comparison.

Plan Limit Awareness

Model and language dropdowns display plan limits when configured. HTTP 402 errors from quota enforcement are surfaced as clear, actionable messages with limit details. Metadata output includes current plan limits for downstream routing.

Use it from an AI Agent, or start from an example

An AI Agent tool

The Entity Enricher node can be attached to an n8n AI Agent as a tool, so an agent decides for itself when to enrich an entity or fetch a schema mid-conversation. The trigger node is deliberately not offered this way — a trigger starts a workflow, it is not something an agent calls.

Six importable workflows

The package ships ready-made workflow files — single enrichment, batch, document to enrichment, image to sample to schema, samples to schema, and draining a delta feed into PostgreSQL. Import one and swap in your own credential rather than wiring it from scratch.

Workflow Ideas

Common patterns for building enrichment pipelines with n8n:

CRM Enrichment— Trigger on new HubSpot/Salesforce contacts, enrich with company data, update the CRM record.

Spreadsheet Pipeline— Read entities from Google Sheets, batch enrich, write results back to a new sheet.

Waterfall Enrichment— Chain multiple enrichment steps with conditional logic — use a fast model first, re-enrich failures with a more capable model.

Scheduled Refresh— Run on a cron schedule to re-enrich stale records with the latest AI models.

Webhook-Driven— Expose a webhook endpoint that accepts entity data, enriches it, and returns the result synchronously.