Quick Start

This guide walks you through the lifecycle of a document: from setting up your workspace to extracting structured data from a file.

Step 1: Check your WorkspaceCopy Anchor Link

A workspace is a secure sandbox. All your documents, Parsings and Operations are isolated within this boundary. Your API key is bound to one workspace and every call acts on it, so paths never name it.

curl "https://api.axelered.com/v1/workspace" \
  -H "Authorization: Bearer <your_api_key>"

Step 2: Upload & Extract DataCopy Anchor Link

An Extraction parses one Document and extracts the fields of your JSON Schema. Send the file in the same call: it becomes a Document of your workspace, kept for 30 days. wait keeps the request open up to 30 seconds for the Extraction to finish.

curl -X POST "https://api.axelered.com/v1/extractions?wait=30" \
  -H "Authorization: Bearer <your_api_key>" \
  -F "file=@/path/to/invoice.pdf" \
  -F "config={
    \"outputSchema\": {
      \"type\": \"object\",
      \"properties\": {
        \"totalAmount\": { \"type\": \"number\" },
        \"vendorName\": { \"type\": \"string\" }
      },
      \"required\": [\"totalAmount\"]
    },
    \"parsing\": { \"tier\": \"accurate\" },
    \"perimeter\": \"external\"
  }"

The response is the Extraction: 200 when it finished within wait, 202 while it is still running. The perimeter says where your data may go: local keeps it inside Axelered's infrastructure, external allows third-party providers. See Perimeters.

Step 3: Read the OutputCopy Anchor Link

Read the Extraction until its status is done. Its output.value carries the extracted fields.

curl "https://api.axelered.com/v1/extractions/{extraction_id}?wait=30" \
  -H "Authorization: Bearer <your_api_key>"

To read the Document itself as Markdown, text, HTML or JSON, submit a Parsing and read its Output. The Parsing reads the Document again and is billed on its own: an Operation never shares its Parsing.

curl -X POST "https://api.axelered.com/v1/parsings?wait=30" \
  -H "Authorization: Bearer <your_api_key>" \
  -F "documentId={document_id}" \
  -F 'config={"tier": "accurate"}'

curl "https://api.axelered.com/v1/parsings/{parsing_id}/output?format=markdown" \
  -H "Authorization: Bearer <your_api_key>"

The documentId is in the Extraction: pass it instead of uploading the file again.