Frequently Asked Questions

Helix Extract is an AI-powered Chrome browser extension that extracts data from PDF, DOCX, and image files and automatically fills web forms. Upload a document, and the AI pulls out the relevant fields. Users can then review the extracted data and fill any browser-based form with one click.

Anyone who regularly transfers data from documents into web forms — customs brokers, freight forwarders, paralegals, insurance processors, HR coordinators, and more.

Helix Extract is a Chrome extension and works in Google Chrome and Chromium-based browsers (Edge, Brave, Arc, etc.).

PDF and DOCX are supported on all plans.

Image files (PNG, JPG/JPEG, GIF, WebP, TIFF) are supported on the Free Trial and Supercharge+ plans.

Yes, the maximum file size is 10 MB per document.

Yes. Scanned PDFs are processed using OCR (optical character recognition) to extract text before AI analysis. Extraction quality depends on scan clarity.

Yes. Multi-page PDFs and DOCX files are fully supported. For longer documents (6+ pages), the system automatically targets the most relevant sections to keep extraction fast and token-efficient.

Yes. Handwriting Mode uses AWS Textract's handwriting detection combined with Claude Vision for multi-pass extraction. Enable it with the "Handwriting Mode" toggle in the extension before uploading.

For best results, use high-resolution scans (300+ DPI) with good lighting and minimal shadows. Handwriting extraction uses more tokens than standard text (~1,500 tokens/page vs. ~500 tokens/page). Handwriting Mode is available on the Free Trial and Supercharge+ plans.

1. Open the extension and upload your document (PDF, DOCX, or image).
2. The extension analyzes the current web page to detect form fields.
3. AI reads the document and extracts matching data fields.
4. Review the extracted data, then click "Fill Form" to populate the page.

Helix Extract uses a two-step pipeline: AWS Textract for OCR (converting document pages to searchable text with bounding-box coordinates), followed by Anthropic's Claude AI models via AWS Bedrock for intelligent field extraction. The system automatically selects the best Claude model for each extraction based on document complexity.

Yes. The Verify Data feature shows a split view with your extracted fields on the left and the original document on the right, with the source text highlighted. You can confirm or edit any value before filling. Verify Data is available on the Free Trial and Supercharge+ plans.

Each extracted field includes a confidence score. Fields below 70% confidence are automatically re-extracted in the background using Claude Opus (a higher-accuracy model) — no action required. Up to 3 low-confidence fields are re-extracted per extraction. On Free Trial and Supercharge+ plans, you can also manually review and edit any value in the Verify Data panel.

Any browser-based web form — government portals (e.g. ACE), CRM systems, case management tools, EHR systems, customs software, and more. If it's a form in your browser, Helix Extract can fill it.

Helix Extract uses a token-based system. Each extraction consumes tokens based on document complexity. Choose from three tiers:

• Free Trial — 15,000 tokens (one-time). Full feature access — no credit card required. Good for roughly 10–15 documents to try the product.
• Premium ($9.99/mo or $7.99/week) — 250,000 tokens/month (62,500/week). PDF & DOCX extraction and form auto-fill. Does not include Verify Data, image uploads, Handwriting Mode, or multi-document upload.
• Supercharge+ ($49.99/mo) — 3,000,000 tokens/month. Full feature access: all Premium features plus Verify Data, image file extraction, Handwriting Mode, multi-document upload, and priority support.

Token usage depends on document type and length:

• Standard typed page: ~500 tokens
• Handwritten page: ~1,500 tokens
• Verify Data (per field): 50 tokens

Example: A 1-page typed PDF with 5 fields uses roughly 600 tokens total.

Both plans include PDF extraction, DOCX extraction, and form auto-fill. Supercharge+ adds:

• Verify Data (review and edit extracted fields before filling)
• Image file extraction (PNG, JPG, WebP, TIFF, GIF)
• Handwriting Mode for handwritten documents
• Multi-document upload
• 12× more monthly tokens (3,000,000 vs. 250,000)
• Priority support

Yes. The Premium plan is also available as a weekly subscription at $7.99/week (62,500 tokens), which resets every 7 days from your subscription start date.

No. Token balances reset each billing cycle for Premium and Supercharge+ tiers. Free-tier tokens do not reset but also do not expire.

Yes. There are no contracts. You can cancel your subscription at any time and retain access until the end of your billing period.

Yes. For high-volume needs, custom token allocations, SLA guarantees, or direct API integration, contact us at info@discoverhelix.com.

All data is encrypted in transit (HTTPS/TLS) and at rest using AWS KMS encryption. Documents are processed and then deleted — they are not stored permanently.

Documents are temporarily uploaded for processing and automatically deleted after extraction is complete. They are not retained.

Extracted data is encrypted and logged for audit purposes only. PII is masked in logs (only partial characters are retained).

We are working toward SOC 2 certification. In the meantime, we employ enterprise-grade encryption, audit logging, and data minimization practices.

Yes. Our Data Processing Agreement outlines our commitments as a data processor, including security measures, sub-processors, and data handling practices. It's designed to support GDPR and enterprise compliance requirements.

1. Create an account at discoverhelix.com.
2. Verify your email.
3. Install the Chrome extension.
4. Start with 15,000 free tokens — no credit card required.

Yes. You can upgrade or downgrade at any time from your account dashboard. Changes take effect immediately with prorated billing applied for the current period.

Your current token balance is displayed in the extension and on your account dashboard.

Make sure the file is under 10 MB and in a supported format. If it's a scanned PDF, ensure the scan is legible. If you're uploading an image or using Handwriting Mode, confirm you're on the Free Trial or Supercharge+ plan — these features are not available on Premium. Try re-uploading or converting to a different file format.

Helix Extract maps extracted data to detected form fields. If a field isn't detected or is mapped incorrectly:

• Click on the form first to ensure the page is fully loaded, then retry.
• Use the Verify Data panel to manually correct values before filling (Free Trial and Supercharge+ only).
• Low-confidence fields are automatically re-extracted in the background — no action needed.
• The extension works on standard web forms but cannot fill forms on browser-internal pages (chrome://, extensions, etc.).

Your current extraction will complete, but you won't be able to start a new one until you have tokens available. Upgrade your plan or wait for your monthly reset.