# Reducto

> Turn PDFs and other documents into clean, structured content with Reducto's document parsing.

App · Document Extraction · 1 action. Web page: https://docs.lodol.com/docs/integrations/reducto

## Actions

### Parse PDF

Parse a PDF (or other supported document) with Reducto and return its extracted content as structured chunks.

**Inputs**

- File (`file`, text, required): The PDF to parse, given by its public URL or uploaded here. Uploaded files are sent to Reducto first, then parsed.

**Outputs**

- `job_id` (text): Reducto job id for this parse
- `result` (object): The parse result. A 'full' result carries the extracted content under 'chunks'; a 'url' result carries a presigned download link when the payload is too large to inline.
- `status` (text)
- `usage` (object): Usage metadata (e.g. num_pages) for the parse

**Example output**

```json
{
  "job_id": "job-abc123",
  "result": {
    "chunks": [
      {
        "content": "# Invoice\n\nInvoice number: INV-2024-001",
        "embed": "Invoice number: INV-2024-001"
      },
      {
        "content": "Total due: $1,234.56",
        "embed": "Total due: $1,234.56"
      }
    ],
    "type": "full"
  },
  "status": "success",
  "usage": {
    "num_pages": 2
  }
}
```