Parsli Documentation

Getting Started

Parsli turns your PDFs, images, invoices, emails, and documents into clean, structured data — automatically. No coding required. Here's how to get up and running in under 5 minutes.

Creating Parsers

A parser is the core building block. Think of it as a "recipe" that tells the AI what data to extract from a specific type of document.

How to create a parser

1. Go to the Parsers page.
2. Click "New Parser".
3. Enter a name and optional description.
4. Click "Create". You'll land on the parser detail page.

Parser status

Each parser has a status that controls whether it processes documents:

Limits

Free accounts can create up to 5 parsers. Paid plans include unlimited parsers.

Defining Your Schema

Your schema tells the AI exactly what data to pull from each document. Each field you add becomes a column in your output data.

Field types

Adding fields

1. Open your parser and go to the Schema tab.
2. Click "Add Field".
3. Enter a field name (e.g. "invoice_number").
4. Select the field type (e.g. "text").
5. Optionally add a description to help the AI understand what to look for.
6. Toggle Required if the field should always be present.
7. Click Save.

Compound fields (tables & nested data)

Use table fields for repeating rows (like invoice line items). Use object fields for grouped data (like a vendor with name + address). Use list fields for simple arrays of values.

Custom extraction instructions

At the bottom of the Schema tab, you'll find a Custom Extraction Instructions text area. Use this to give the AI parser-level guidance that applies to every extraction.

Testing Extraction

Before connecting integrations, test your parser to make sure it extracts the right data.

1. Open your parser and go to the Test tab.
2. Drag-and-drop a sample document (or click to browse).
3. Wait a few seconds while the AI processes your document.
4. Review the extracted data in Table view or JSON view.
5. If results aren't right, go back to Schema and adjust your fields or instructions.

Reading your results

The results view shows each field with its extracted value. Pay attention to:

• Table view — shows fields in a readable format with type badges
• JSON view — raw output, useful for developers to verify structure
• Missing values — shown as "-" when the AI couldn't find data
• Confidence — the AI assigns confidence scores; low-confidence fields are flagged

Integrations

Once your parser is working, connect it to the tools you already use. Open your parser's Integrations tab to add any of the following:

Google Sheets

Pull extracted data directly into a Google Sheet — no code needed. Parsli generates a live data feed URL that Google Sheets can read automatically.

1. In the Integrations tab, click "Add Integration" and select Google Sheets.
2. Copy the generated IMPORTDATA formula.
3. Paste it into any cell in your Google Sheet.
4. Your sheet now automatically pulls the latest extracted data as CSV.

=IMPORTDATA("https://parsli.co/api/parsers/{id}/feed?token={token}&format=csv")

Zapier

Connect to 5,000+ apps through Zapier. Every time a document is processed, the extracted data is sent to your Zap.

1. In Zapier, create a new Zap with the "Webhooks by Zapier" trigger.
2. Choose "Catch Hook" and copy the webhook URL.
3. In Parsli, add a Zapier integration and paste the webhook URL.
4. Click "Test" to send a sample payload to Zapier.
5. In Zapier, continue building your Zap with the extracted data fields.

Make (formerly Integromat)

Similar to Zapier — create a Make scenario with a "Custom webhook" module, copy the URL into Parsli, and your extracted data flows into your automation.

Custom Webhooks

Send extracted data to any HTTP endpoint. Perfect for developers who want to push data to their own backend, database, or custom application.

1. Add a Webhook integration.
2. Enter your endpoint URL and HTTP method (POST or PUT).
3. Optionally configure authentication (Bearer token or Basic auth).
4. Click "Test" to verify delivery.

Webhook payload format:

{
  "event": "document.processed",
  "parser_id": "abc-123",
  "parser_name": "Invoice Parser",
  "document_id": "doc-456",
  "timestamp": "2026-03-06T12:00:00Z",
  "data": {
    "invoice_number": "INV-2026-001",
    "vendor_name": "Acme Corp",
    "total_amount": 1250.00,
    "due_date": "2026-04-01"
  },
  "metadata": {
    "file_name": "invoice.pdf",
    "mime_type": "application/pdf",
    "source_type": "upload",
    "page_count": 1
  }
}

Gmail Inbox

Automatically extract data from email attachments. Parsli connects to your Gmail account (read-only) and processes attachments from specific senders.

1. Add a Gmail Inbox integration.
2. Click "Connect Gmail Account" and authorize read-only access.
3. Set a from filter — only emails from this address will be processed (e.g. invoices@supplier.com).
4. Click "Activate".

Parsli checks your inbox every 5 minutes. When a matching email with an attachment arrives, it's automatically extracted and sent to your other integrations (webhooks, Google Sheets, etc.).

API Reference

For developers who want to integrate extraction into their own applications. The API accepts documents and returns structured JSON.

Authentication

API requests require a Bearer token. Create API keys in your parser's API tab.

1. Open your parser and go to the API tab.
2. Click "Create Key" and give it a name.
3. Copy the key immediately — it's shown only once.
4. Store it securely (environment variable, secret manager, etc.).

Extract document

Send a document for extraction using the REST API:

curl -X POST https://parsli.co/api/v1/extract \
  -H "Authorization: Bearer ext_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "file": {
      "name": "invoice.pdf",
      "type": "application/pdf",
      "data": "BASE64_ENCODED_FILE_CONTENT"
    }
  }'

Response:

{
  "success": true,
  "parser_id": "abc-123",
  "document_id": "doc-456",
  "results": {
    "invoice_number": "INV-2026-001",
    "vendor_name": "Acme Corp",
    "total_amount": 1250.00,
    "due_date": "2026-04-01",
    "line_items": [
      { "description": "Widget A", "quantity": 10, "unit_price": 100.00, "total": 1000.00 },
      { "description": "Widget B", "quantity": 5, "unit_price": 50.00, "total": 250.00 }
    ]
  }
}

Inbound webhook

Each parser also has a unique inbound webhook URL for sending documents without API keys. Find it in the API tab. Send documents as multipart form data or JSON with base64 content.

# Using multipart form data
curl -X POST https://parsli.co/api/inbound/webhook/YOUR_TOKEN \
  -F "file=@invoice.pdf"

Error codes

Credits & Billing

Parsli uses a simple page-based credit system. Each document extraction uses 1 credit, regardless of file type or size.

How credits work:

• Credits reset automatically every 30 days.
• 1 document = 1 credit, regardless of file size or number of pages.
• Test extractions also use credits.
• If you run out of credits, extractions will pause until your next billing cycle.
• Your usage is visible on the dashboard and in Settings.

Supported File Types

Frequently Asked Questions