codemation-document-scanner

name	codemation-document-scanner
description	CodemationDocumentScanner node — managed document/invoice/image extraction via the Codemation doc-scanner service. No Azure credentials required. Read before writing any workflow that scans documents, invoices, or images.
compatibility	Codemation core-nodes. Requires @codemation/core-nodes import.
tags	ocr, document, invoice, image, scan, extract, managed, confidence
uses	@codemation/core-nodes

Codemation Document Scanner Node

Start here: call find_examples before reading further.

• find_examples({ query: "CodemationDocumentScanner" }) — node-level usage and all analyzerType variants
• find_examples({ query: "invoice scan post accounting" }) — end-to-end invoice extraction scenario
• find_examples({ query: "document scanner confidence fields" }) — how to enable per-field confidence scores

Use this skill when

Writing a workflow that extracts text and/or structured fields from documents, invoices, or images using the Codemation managed scanning service. No Azure credentials are required — the service is pre-wired to the platform.

Use codemation-workflow-dsl for surrounding workflow structure. Use codemation-ai-agent-node if you need to pass the extracted markdown to an LLM for further processing.

When to use CodemationDocumentScanner vs the standalone OCR nodes

Situation	Use
Managed platform deployment, no Azure credential	codemationDocumentScannerNode (this skill)
Self-hosted / BYOK Azure Content Understanding	analyzeDocumentNode / analyzeInvoiceNode / analyzeImageNode from @codemation/core-nodes-ocr

codemationDocumentScannerNode calls the internal doc-scanner service via HMAC — the workspace holds no Azure key. The standalone OCR nodes call Azure directly using a per-workspace credential.

Choosing analyzerType

analyzerType	When to use	Azure analyzer	Field extraction	Confidence opt-in supported
"document"	General PDFs, Word docs, HTML, text-heavy files	prebuilt-document	Yes	Yes
"invoice"	Invoices, receipts — always prebuilt-invoice	prebuilt-invoice	Yes	Yes
"image"	Photos, screenshots, diagrams	prebuilt-imageAnalyzer	No (markdown only)	No — image carries no extraction charge
"auto"	Unknown mime type at author time	Routes on Content-Type: image/* → image, everything else → document	Depends on routed type	Depends on routed type

Default is "auto". Set an explicit type whenever you know the content class — it avoids unnecessary re-routing and makes the workflow self-documenting.

Output shape

{
  markdown: string; // full text content
  fields: Record<
    string,
    {
      value: unknown; // extracted scalar, date ISO string, nested object, or array
      confidence: number | null; // 0–1 when includeConfidence:true; null otherwise
    }
  >;
}

item.json.markdown is the Markdown rendering of the document. item.json.fields is a flat-or-nested map of structured fields found by the analyzer. For analyzerType: "image", fields is always {}. Fields may be sparse or absent for generic documents — extraction is best-effort.

WARNING: How to enable per-field confidence scores (LD6)

By default, confidence is null on every field. This keeps token cost low for the majority of workflows that only need value.

To get a populated confidence (0–1 float) on each field, set includeConfidence: true:

codemationDocumentScannerNode.create(
  {
    binaryField: "data",
    analyzerType: "invoice",
    includeConfidence: true, // ← opt-in: fields carry confidence 0–1
  },
  "Scan invoice",
  "scan-invoice",
);

Cost implication: enabling confidence routes the request to a confidence-enabled analyzer variant, which roughly doubles the contextualization token count for document/invoice analyzers. Only enable it when your downstream logic actually reads field.confidence.

Images (analyzerType: "image") and auto-routed-to-image requests carry no field-extraction charge regardless of this flag — they silently ignore includeConfidence (confidence stays null, never a 400).

API usage

import { codemationDocumentScannerNode } from "@codemation/core-nodes";

codemationDocumentScannerNode.create(
  {
    binaryField?: string;         // key on item.binary — default "data"
    analyzerType?: "document" | "invoice" | "image" | "auto";  // default "auto"
    contentType?: string;         // MIME type override — falls back to attachment.mimeType
    includeConfidence?: boolean;  // default false — see cost note above
    maxBytes?: number;            // size cap before reading; default 50 MiB (LD10)
  },
  label?: string,   // node label on the canvas
  nodeId?: string,  // explicit stable id — set when output is used downstream
)

Set an explicit nodeId whenever downstream nodes reference this node's output by id, or when the node may be renamed later (avoids credential-binding orphaning).

Consuming fields downstream

import type { DocScannerOutput, DocScannerField } from "@codemation/core-nodes";

// item.json after codemationDocumentScannerNode:
// DocScannerOutput = { markdown: string; fields: Record<string, DocScannerField> }
// DocScannerField  = { value: unknown; confidence: number | null }

new Callback<DocScannerOutput>("Use fields", (items, _ctx) =>
  items.map((item) => {
    const vendorName = item.json.fields["VendorName"]?.value as string | undefined;
    const vendorConf = item.json.fields["VendorName"]?.confidence; // null unless includeConfidence:true
    return { ...item, json: { ...item.json, vendorName, vendorConf } };
  }),
);

Read next when needed

• codemation-workflow-dsl — workflow builder, trigger types, fluent vs low-level API.
• codemation-ai-agent-node — pass item.json.markdown to an LLM for summarisation or extraction.