Foundry API DocuStruct Extract: JSON from Documents in n8n

Extract structured JSON from documents with Foundry—directly inside n8n

This n8n workflow runs manually to send a Base64-encoded document plus a text snippet to Foundry’s DocuStruct Extract and JSON Forge endpoints—then returns normalized, schema-aligned JSON for fast downstream automation.

What this workflow does

Designed for n8n users, automation engineers, and SaaS operators, it performs two coordinated extraction/normalization steps:

• Manual trigger: The workflow starts when you run it manually in n8n.
• Set your input payload: You provide filename, mimeType, the document as Base64 file content, a target JSON schema, and an optional text string.
• Foundry DocuStruct Extract API call: The workflow sends your Base64 document and the schema to extract structured JSON from the document.
• Foundry JSON Forge API call: It then normalizes the provided text into structured JSON that matches the same schema.

Use cases

• Document-to-JSON ingestion: Convert uploaded documents into consistent, schema-ready JSON for indexing, review, or storage.
• Hybrid extraction & normalization: Use document extraction (DocuStruct Extract) alongside text normalization (JSON Forge) to standardize outputs.
• Schema-driven workflows: Apply a custom schema to ensure extracted/normalized data stays compatible with your existing systems.

Technical details

• Authentication: Create an n8n HTTP Header Auth credential that sends your Foundry API key as X-API-Key, and select it in both Foundry HTTP Request steps.
• Nodes used: Manual Trigger, Set, Sticky Note, and HTTP Request.
• Setup: Replace REPLACE_WITH_AUTHORIZED_BASE64_FILE with your document’s Base64 content and set correct filename and mimeType. Update the schema (and optional text) to match the structure you want Foundry to extract/normalize.